This article has been accepted for publication in a future issue of this journal, but has not been
fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
Towards Smart Wireless Communications via Intelligent
Reflecting Surfaces: A Contemporary Survey
Shimin Gong, Member, IEEE, Xiao Lu, Member, IEEE, Dinh Thai Hoang, Member, IEEE, Dusit Niyato, Fellow,
IEEE, Lei Shu, Senior Member, IEEE, Dong In Kim, Fellow, IEEE, and Ying-Chang Liang, Fellow, IEEE
Abstract—This paper presents a literature review on recent
applications and design aspects of the intelligent reflecting sur-
face (IRS) in the future wireless networks. Conventionally, the
network optimization has been limited to transmission control
at two endpoints, i.e., end users and network controller. The
fading wireless channel is uncontrollable and becomes one of
the main limiting factors for performance improvement. The
IRS is composed of a large array of scattering elements, which
can be individually configured to generate additional phase
shifts to the signal reflections. Hence, it can actively control
the signal propagation properties in favor of signal reception,
and thus realize the notion of a smart radio environment. As
such, the IRS’s phase control, combined with the conventional
transmission control, can potentially bring performance gain
compared to wireless networks without IRS. In this survey, we
first introduce basic concepts of the IRS and the realizations of
its reconfigurability. Then, we focus on applications of the IRS
in wireless communications. We overview different performance
metrics and analytical approaches to characterize the perfor-
Manuscript received December 16, 2019; revised February 21, 2020 and
May 10, 2020; accepted June 16, 2020. The work of Shimin Gong was
supported in part by National Natural Science Foundation of China (NSFC)
under Grant 61972434, Shenzhen Basic Research Program under Grant
JCYJ20190807154009444, Shenzhen Talent Peacock Plan Program under
Grant KQTD2015071715073798, and PCL Future Greater-Bay Area Network
Facilities for Largescale Experiments and Applications (LZC0019). The work
of Dusit Niyato was supported in part by the National Research Foundation
(NRF), Singapore, under Singapore Energy Market Authority (EMA), Ener-
gy Resilience, NRF2017EWT-EP003-041, Singapore NRF2015-NRF-ISF001-
2277, Singapore NRF National Satellite of Excellence, Design Science and
Technology for Secure Critical Infrastructure NSoE DeST-SCI2019-0007,
A*STAR-NTU-SUTD Joint Research Grant on Artificial Intelligence for the
Future of Manufacturing RGANS1906, Wallenberg AI, Autonomous Systems
and Software Program and Nanyang Technological University (WASP/NTU)
under grant M4082187 (4080). The work of Dong In Kim was supported by
the National Research Foundation of Korea (NRF) Grant funded by the Korean
Government (MSIT) under Grant 2014R1A5A1011478. The work of Ying-
Chang Liang was supported in part by the National Natural Science Founda-
tion of China under Grant U1801261, the Key Areas of Research and Develop-
ment Program of Guangdong Province, China, under Grant 2018B010114001,
the Central Universities Funding under Grant ZYGX2019Z022, and the 111
Project under Grant B20064. Corresponding author: Ying-Chang Liang.
S. Gong is with School of Intelligent Systems Engineering, Sun Yat-
sen University, Guangzhou 510275, China, and also with Peng Cheng
Laboratory, Shenzhen 518055, China (e-mail: gongshm5@mail.sysu.edu.cn).
X. Lu is with the Department of Electrical and Computer Engineering,
University of Alberta, Edmonton, AB, Canada (e-mail: lu9@ualberta.ca).
D. T. Hoang is with the Faculty of Engineering and Information Technology,
University of Technology Sydney, Ultimo, NSW 2007, Australia (e-mail:
hoang.dinh@uts.edu.au). D. Niyato is with School of Computer Science and
Engineering, Nanyang Technological University, Singapore (e-mail: dniya-
to@ntu.edu.sg). L. Shu is with NAU-Lincoln Joint Research Center of Intel-
ligent Engineering, Nanjing Agricultural University, Najing 210095, China (e-
mail: lei.shu@njau.edu.cn). D. I. Kim is with the Department of Electrical and
Computer Engineering, Sungkyunkwan University (SKKU), Suwon, South
Korea (e-mail: dikim@skku.ac.kr). Ying-Chang Liang is with the Center for
Intelligent Networking and Communications (CINC), University of Electronic
Science and Technology of China, Chengdu 611731, China, and also with
Shenzhen Institute of Advanced Studies, University of Electronic Science and
Technology of China, Shenzhen, China (e-mail: liangyc@ieee.org).
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
mance improvement of IRS-assisted wireless networks. To exploit
the performance gain, we discuss the joint optimization of the
IRS’s phase control and the transceivers’ transmission control
in different network design problems, e.g., rate maximization
and power minimization problems. Furthermore, we extend the
discussion of IRS-assisted wireless networks to some emerging
use cases. Finally, we highlight important practical challenges
and future research directions for realizing IRS-assisted wireless
networks in beyond 5G communications.
Keywords- Intelligent reflecting surface, smart radio environ-
ment, passive beamforming, IRS-assisted wireless networks.
I. INTRODUCTION
With the popularizing of user devices constituting the future
Internet of Things (IoT), we have never stopped our efforts on
the challenging network optimization problems to improve the
energy- or spectrum-efficiency (EE/SE) of wireless networks,
with the aim of accommodating the users’ demanding data rate
and diverse quality of service (QoS) requirements, e.g., [1]
and [2]. Currently, the performance optimization of wireless
networks either focuses on the user side or the network
controller, e.g., the base station (BS) and network operator.
For wireless network operators, the ever-increasing traffic
demand can be fulfilled by deploying energy-efficient small
cells in a dense network or using multiple antennas at the
BS to increase spectrum efficiency [3]. The BS’s transmit
beamforming or power allocation can be optimized to adapt
to the channel variations. At the user side, multiple users
can join collaboration, e.g., via device-to-device (D2D) [4]
and relay communications [5]. These features can potentially
provide the benefits of improved link quality and coverage,
increased EE/SE performance, reduced interference and power
consumption [6]. A joint optimization can be made possible
when the information exchange and coordination between end
users and the network controller are available. This is preferred
as it generally yields a higher performance gain if it is solvable
with affordable cost. Hence, numerous research works in the
literature have proposed joint system optimizations to improve
the EE/SE performance of wireless networks by a combination
of different techniques, e.g., [7] and [8]. These may include the
optimization for wireless power transfer, cooperative relaying,
beamforming, and resource allocation, etc.
However, in the current paradigm of wireless network
optimization, the radio environment itself remains an uncon-
trollable factor and thus it is not accounted for in the problem
formulations. Due to randomness in the radio environment, the
signal propagation typically experiences reflections, diffrac-
tions, and scattering before reaching the receiver with multiple
randomly attenuated and delayed copies of the original signals
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

Active RF communications Passive signal reflections

Intelligent Reflecting Backscatter transmitter

Base station Surface (IRS) AF Relay
(BTx)
(BS)

h s[n] h s[n] h s[n]

IRS controller

gΘh s[n] gh s[n]b[k] g𝛽hs[n]

f s[n] f s[n] f s[n]
Backscatter
Active receiver receiver (BRx) Active receiver

(a) IRS-assisted wireless communications (b) Wireless backscatter communications (c) AF-Relay assisted wireless communications
to deliver s[n] from the BS to deliver b[k] from the BTx to deliver s[n] from the BS

Fig. 1: Comparing IRS-assisted wireless communications with the backscatter communications and amplify-and-forward (AF)
relay-assisted wireless communications. The IRS in (a) introduces a phase shift matrix Θ to configure the equivalent reflecting
channel. The AF relay in (c) introduces a power amplifying coefficient β to forward the received signal. The receiver decodes the
n-th source information symbol s[n] for the IRS-assisted and AF relay-assisted communications, while in (b) it aims at decoding
the k-th piggybacked information symbol b[k] from the strong interference s[n] in wireless backscatter communications. The
BS-IRS, BS-receiver, and IRS-receiver channels are denoted by h, f , g, respectively.

in different paths. Such a channel fading effect becomes a
major limiting factor for the maximization of EE/SE perfor-
mance of wireless networks. Recently, a novel concept of
intelligent reflecting surface (IRS) has been introduced in
the wireless communications research community [9]–[11].
The IRS is a two-dimensional (2D) man-made surface of
electromagnetic (EM) material, namely metasurface, that is
composed of a large array of passive scattering elements with
specially designed physical structure.1 Each scattering element
can be controlled in a software-defined manner to change
the EM properties (e.g., the phase shift) of the reflection of
the incident RF signals upon the scattering elements. By a
joint phase control of all scattering elements, the reflecting
phases and angles of the incident RF signals can be arbitrarily
tuned to create a desirable multi-path effect. In particular, the
reflected RF signals can be added coherently to improve the
received signal power or combined destructively to mitigate
interference. A typical system model of IRS-assisted wireless
communications is illustrated in Fig. 1(a). By deploying the
IRS in the environment, e.g., coated on walls of buildings
and carried by aerial platforms, the IRS can turn the radio
environment into a smart space that can assist information
sensing, analog computing, and wireless communications [10].
Along with the optimal control of the legacy RF transceivers,
IRS-assisted wireless systems will become more flexible to
support diverse user requirements, e.g., enhanced data rate,
extended coverage, minimized power consumption, and more
secure transmissions [12]–[14].
The vision of smart radio environment has been demon-
strated via realistic simulations and experiments to verify its
capability of improving transmission performance in different
wireless networks leveraging the IRS’s reconfigurability. The
1 In the literature, we have observed different names referring to a similar
concept of using reconfigurable metasurface to assist wireless communica-
tions, e.g., software-defined hypersurface, large intelligent surface/antenna,
reconfigurable intelligent surface, and holographic MIMO surface. For con-
sistence, we use the term of intelligent reflecting surface (IRS) throughout
this paper.
active wall is introduced in [15], [16] by using an active
frequency-selective surface (FSS) to manipulate the wireless
environment. The FSS provides a narrow-band frequency
filtering of incoming signals [24], which can be used to build
up a cognitive engine that makes the walls intelligent. The
experiments in [18] demonstrate that, by deploying hybrid
active-passive elements in the walls of a building, we can
expect to 1) mitigate the destructive effect of multiple paths for
legacy wireless communications, 2) eliminate poorly condi-
tioned similar channels in a large-scale MIMO system, and 3)
simultaneously attenuate interference power and increase sig-
nal strength at different receivers. The idea of programmable
wireless channels or environment is proposed in [19], [20] by
using hypersurfaces, i.e., software-controlled metamaterials,
to cover physical objects in the radio environment. Several
physical layer building-block technologies for a programmable
wireless environment are implemented and evaluated in [25].
By controlling the distribution of the current over the hy-
persurfaces, the incident RF signals can be reshaped with a
desirable response, thus realizing a reconfigurable wireless
environment. By alleviating signal path loss, multi-path fading,
and co-channel interference, the programmable wireless en-
vironment is capable of improving transmission performance
in terms of signal quality, communication range, and EE/SE
performance. Different from the use of hypersurfaces, the
authors in [21] introduce Ultra-Massive MIMO platforms to
realize an intelligent communication environment, which en-
ables full-wave control with plasmonic arrays deployed at the
transmitter, through the channel, and at the receiver. Although
massive MIMO can provide remarkable improvement to the
wireless channel conditions, the EM manipulation of massive
MIMO is only feasible at the transceivers, while the wireless
environment remains passive and uncontrollable. As a result,
the channel model is still a probabilistic process, rather than
a software-defined reconfigurable counterpart that is actively
participating in the communication process. A summary of
different concepts and realizations for a smart radio envi-
ronment is listed in Table I. The application scenarios are

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

TABLE I: Using IRS for Realizing the Vision of Smart Radio Environment

Reference Concept/Scheme Design approach Application scenario Validation

[15] Intelligent wall Switch FSS between ON and OFF to Smart indoor environment Extend coverage and improve system
shape the propagation environment for OFDMA system performance up to 80% by simulation
[16] ANN-based Use ANN to explore the optimal setting Smart indoor environment Simulations show quick responses to
intelligent wall for controlling the intelligent walls at 2 GHz demands and improved performance
[17] Spatial microwave Use a binary-phase state tunable meta- SMM fabricated at the Experiments show the capability of im-
modulator (SMM) surface to manipulate EM waves 2.47 GHz frequency proving or cancelling RF signals
[18] Programmable radio Embed low-cost devices in walls to Indoor 2.4 GHz Wi-Fi-like Experiments show the efficacy of atten-
environment passively reflect active RF signals communications uating or enhancing signal by 26 dB
[19] Programmable wire- Control current distribution over hyper- Indoor 60 GHz mmWave Simulations demonstrate significantly
less environment surface tiles to manipulate EM waves communications improved coverage and received power
[20] Programmable wire- Hypersurface for interference control, Indoor 2.4 GHz and 60 Simulations show groundbreaking per-
less environment security, and distortion mitigation GHz communications formance and security potential
[21] Ultra-Massive MIMO Deploy plasmonic arrays at the mmWave and THz com- Significant improvements in communi-
transceivers and through the channel munications cation distance and data rate
[22] Intelligent mirror Create LOS link by rotating the IRS or Free space optical (FSO) Simulations show that building sway
electronically changing the wavefront communications for the IRS has either a smaller or larger
impact on the channel quality
[23] Physical shaping of Install binary-phase state tunable meta- 2.47 GHz Wi-Fi frequency Experiments show perfect channel or-
propagation medium surface inside random environment thogonality, optimal channel diversity,
flexible interference suppression

rather diverse, from media sharing Wi-Fi systems to mmWave,
THz, and even optical communications, covering a broadband
frequency range, while the common observation is that the
IRS can be utilized to create desirable radiation patterns in
favor of wireless communications.
By integrating the smart radio environment into the network
optimization problems, IRS-assisted wireless networks are
envisioned to revolutionize the current network optimization
paradigm and expected to play an active role in future wire-
less networks [3], [26]. The specific benefits of IRS-assisted
wireless communications can be summarized as follows:
• Easy deployment and sustainable operations: The IRS
is made of low-cost passive scattering elements embedded
in the metasurface. It can be in any shape, thus providing
high flexibility for its deployment and replacement. It
can be easily attached to and removed from facades of
buildings, indoor walls, and ceilings, etc. Without the
use of active components for power-consuming signal
processing algorithms, the IRS can be battery-less and
wirelessly powered by RF-based energy harvesting.
• Flexible reconfiguration via passive beamforming: The
IRS can bring additional phase shifts to the reflected
signals. By jointly optimizing the phase shifts of all scat-
tering elements, namely passive beamforming, the signal
reflections can be coherently focused at the intended
receiver and nulled at other directions. The number of
reflecting elements can be extremely large, e.g., from
tens to hundreds [12], depending on the traverse size of
the IRS. This implies a great potential for performance
improvement of wireless networks. The IRS’s phase con-
trol, combined with the transceivers’ operational param-
eters, e.g., transmit beamforming, power allocation, and
resource allocation, can be jointly optimized to explore
the performance gain of IRS-assisted wireless networks.
• Enhanced capacity and EE/SE performance: By us-
ing the IRS, the wireless channel can be programmed
to support a higher link capacity with reduced power
consumption for point-to-point communications. Interfer-
ence suppression also becomes effective by using the
IRS, which implies a better signal quality for the cell-
edge users. For multi-user (MU) wireless networks, the
scattering elements can be partitioned and allocated to
assist data transmissions of different users. As such,
the IRS-assisted wireless network can provide better
QoS provisioning and potentially improve the sum-rate
performance or max-min fairness among different users.
• Exploration of emerging wireless applications: The
development of the IRS is expected to pave the way
for new promising research directions. For example, the
IRS has recently been introduced to be a novel approach
for preventing wireless eavesdropping attacks by simul-
taneously controlling the transmission at the transmitter
and the reflections at the IRS. Many other emerging
research areas also benefit from the use of the IRS such as
wireless power transfer, unmanned aerial vehicle (UAV)
communications, and mobile edge computing (MEC),
which will be reviewed in this survey.
Although the IRS’s operation resembles that of a multi-
antenna relay, it is fundamentally different from the existing
relay communications. By using passive elements, the IRS
can realize fully controllable beam steering without dedicated
energy supply and sophisticated active circuitry for channel
estimation, information decoding, and amplifying and forward-
ing. Compared to the conventional amplify-and-forward (AF)
relay that actively generates new RF signals, the IRS does
not use an active transmitter but only reflects the ambient
RF signals as a passive array, which incurs no additional
power consumption. It also differs from the conventional
backscatter communications, where the backscatter transmitter
communicates with its receiver by modulating and reflecting
the ambient RF signals [27]. The backscattered information is
piggybacked in the ambient RF signals. Despite their differ-
ences, the IRS can also be used to perform wireless backscatter
communications due to its capability of manipulating the

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

Toward Smart Wireless Communications via Intelligent Reflecting Surfaces
ducing the hardware architectures, classifications, and main
II. THEORY AND DESIGN OF INTELLIGENT REFLECTING  characteristics. Several short survey and tutorial papers also
SURFACES appear in the literature, e.g., [11], [14], mainly discussing the
• The IRS' Structure and Control Mechanism recent applications of the IRS in typical wireless scenarios.
• Typical Tunable Functions The tutorial in [11] covers an overview of the IRS technology
• Experiments and Prototypes
to create smart and reconfigurable environment, including
III. PERFORMANCE ANALYSIS OF IRS‐ASSISTED WIRELESS  its main applications in wireless communication, competitive
NETWORKS
advantages over existing technologies, hardware architecture
• IRS‐enhanced Path Loss Models as well as the corresponding new signal model, with the focus
• Using IRS as the Signal Reflector
• Using IRS as the Signal Receiver on the key challenges in designing and implementing the IRS-
• Using IRS as the Signal Transmitter assisted wireless networks. Liang et al. in [14] firstly gave a
IV. APPLICATION AND OPTIMIZATION OF IRS‐ASSISTED 
tutorial on the evolving from the backscatter communications,
WIRELESS NETWORKS reflective relay, to intelligent reflective surfaces, which are all
relying on the use of inexpensive and passive components to
• SNR or Capacity Maximization
• Power Minimization or EE/SE Maximization realize enhanced wireless information transmission by using
• IRS‐assisted Physical Layer Security EM scattering principles. The tutorial covers a comparison
V. EMERGING USE CASES OF IRS IN WIRELESS NETWORKS
of the signal models and design aspects of different radio
technologies. Comparing to these works, our survey offers
• Deep Learning for IRS‐assisted Systems a comprehensive study on the theoretical basis of the IRS
• IRS‐assisted Wireless Power Transfer
• IRS‐assisted UAV Communications and a contemporary review on its most recent applications
• IRS‐assisted Mobile Edge Computing in wireless networks. Specifically, our major contributions are
summarized as follows:
VI. CHALLENGES AND FUTURE RESEARCH DIRECTIONS
• A systematic organization of the literature under exten-
Fig. 2: Survey framework and outline of the main topics.
sive review is provided (as shown in Fig. 2) for readers
with different backgrounds and interests to comprehend
the IRS technology more effectively. We start from the
phases of the reflected signals, e.g., [28]–[30]. The evolving
physical characterization of the IRS and its EM proper-
path from backscatter radios to large intelligent surfaces is
ties, covering both design methodologies and experimen-
revealed in [14]. The difference among three communication
tal prototypes. Then we focus on IRS-assisted wireless
technologies is illustrated in Fig. 1. Such a different feature
networks by classifying different papers according to their
of IRS-assisted wireless networks thus motivates the necessity
design objectives and control variables.
of presenting a comprehensive literature review, which aims
• We provide an extensive review on the stochastic analysis
at providing fundamental knowledge about the IRS’s physical
of performance limit and asymptotic behavior of IRS-
properties and implementations, diverse applications in differ-
assisted wireless networks, which are not covered by
ent network scenarios, and the performance optimizations of
the existing review papers. The pervasive deployment of
IRS-assisted wireless networks.
the IRS will change the random nature of the channel
There are a few review papers in the literature, either
environment, which calls for different analytical tools and
focusing on the physical design and implementations of the
performance metrics to characterize the performance lim-
reconfigurable metasurface, e.g., [31]–[33], or its conceptu-
its. Corroborated by the stochastic analysis, the potential
al applications in wireless communications and networking,
increase in performance gain then motivates the further
e.g., [10], [12], [13]. In particular, the review papers [31]–[33]
optimization of IRS-assisted wireless networks.
mainly focus on the theoretical basis, physics characterization,
• Some common shortcomings of the current literature are
and classification of metasurfaces as well as their applications
analyzed and technical insights are highlighted for future
at different operational frequencies. Motivated by the IRS’s
exploration. For example, the current optimization frame-
appealing EM properties, the authors in [10] discuss the feasi-
works for IRS-assisted wireless networks mainly rely on
bility and methodologies of using the IRS to realize the con-
the alternating optimization method, which implies that
cept of smart radio environments, and also analyze potential
a higher performance gain can be achieved potentially
solutions to some fundamental challenges towards its massive
with a more sophisticated algorithm design. Besides, a
deployment and applications in future wireless networks. Fol-
majority of the papers in the literature focus on joint
lowing on, the review paper in [12] provides a more technical
active and passive beamforming optimization. In fact,
overview with a special focus on the analysis of signal models
the passive beamforming can also be jointly optimized
and the physical layer channel enhancement in IRS-assisted
with other control strategies, e.g., information encoding,
wireless communications. The authors in [13] introduce the
transmission scheduling, access control, and full-duplex
concept of holographic multiple-input multiple-output (MI-
communications. We also notice that the energy con-
MO) surface (HMIMOS) comprising sub-wavelength metallic
sumption of the IRS is usually omitted in the literature,
or dielectric scattering particles, which shares a similar idea
which may lead to over-optimistic conclusions.
with the use of the IRS in wireless communications. An
overview of HMIMOS communications is provided by intro- The rest of this paper is organized as follows. Section II

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

TABLE II: List of abbreviations II. THEORY AND DESIGN OF INTELLIGENT

Abbreviation Description REFLECTING SURFACES
3D/2D 3-Dimensional/2-Dimensional The metasurface is a kind of two-dimensional (i.e., with
ADMM Alternating Direction Method of Multipliers near-zero thickness) man-made material that exhibits special
AF/DF Amplify/Decode-and-Forward EM properties depending on its structural parameters. As
AI/ANN Artificial Intelligence/Artificial Neural Network
illustrated in Fig. 3, the metasurface is composed of a large
AP/BS Access Point/Base Station
array of passive scattering elements, e.g., metallic or dielectric
particles, that can transform the impinging EM waves in dif-
BCD Block Coordinate Descent
ferent ways [34]. The sub-wavelength structural arrangement
BER/SER Bit Error Rate/Symbol Error Rate
of the scattering elements determines how the incident waves
CSI/CRN Channel State Information/Cognitive Radio Network are transformed, i.e., the direction and strength of the reflected
D2D/DC/DL Device-to-Device/Difference-of-Convex/Deep Learning and diffracted waves. In general, when EM waves propagate
DNN/DRL Deep Neural Network/Deep Reinforcement Learning to a boundary between two different medias, the strengths and
EE/SE/EM Energy-Efficiency/Spectrum-Efficiency/Electromagnetic directions of the reflected and diffracted waves typically follow
FPGA/FSS Field-Programmable Gate Array/Frequency-Selective Surface the Fresnel equations and Snell’s law, respectively [32]. The
GHz/THz Gigahertz/Terahertz situation becomes different when the same wave impinges up-
HetNets Heterogeneous Networks
on a metasurface. The periodical arrangement of the scattering
IC/NoC Integrated Circuit/Network-on-Chip
elements can cause a shift of the resonance frequency and thus
a change of boundary conditions. As a result, the reflected and
IoT/IRS Internet of Things/Intelligent Reflecting Surface
diffracted waves will carry additional phase changes.
LOS/LT/LR Line-of-Sight/Legitimate Transmitter/Legitimate Receiver
Once the metasurface is fabricated with a specific physical
MAC/MEC Medium Access Control/Mobile Edge Computing
structure, it will have fixed EM properties and therefore can be
MISO/MIMO Multiple-Input and Single-Output/Multiple-Output used for a specific purpose, e.g., a perfect absorber operating
MM/MU Majorization Minimization/Multi-user at a certain frequency. The analysis of EM properties can
mmWave Millimeter Wave be based on the general-purpose full-wave EM simulator or
OFDM Orthogonal Frequency Division Multiplexing approximate computational techniques [35]. More efficient
OFDMA Orthogonal Frequency Division Multiple Access analytical approaches rely on sophisticated boundary condi-
QoS Quality of Service
tions to describe the metasurface discontinuity and its EM
responses [36], [37]. However, it becomes very inflexible as
RB/RF/RL Resource Block/Radio Frequency/Reinforcement Learning
a new metasurface has to be re-designed and fabricated to
SCA/SDR Successive Convex Approximation/Semidefinite Relaxation
serve another purpose or operate at a different frequency. In
SNR/SINR Signal-to-Noise Ratio/Signal-to-Interference plus Noise Ratio
particular, based on the application requirements, the structural
SM/SMM Spatial Modulation/Spatial Microwave Modulator parameters of the scattering elements constituting the metasur-
SOCP Second-Order Cone Programming face have to be recalculated by a synthesis approach [38], [39],
SWIPT Simultaneous Wireless Information and Power Transfer which is in general computational demanding.
TDMA/NOMA Time Division/Non-Orthogonal Multiple Access The IRS is built from a reconfigurable metasurface, which
UAV Unmanned Aerial Vehicle can fully control the phase shifts incurred by individual
ZF Zero-Forcing scattering elements. This can be achieved by imposing external
stimuli on the scattering elements and thus alter their physical
parameters, leading to the change of EM properties of the
metasurface without refabrication [40]. The first design issue
for the IRS lies in a control mechanism to connect and
communicate with a large size of scattering elements, and thus
presents the basic theory and implementation of the IRS. agilely and jointly control their EM behaviors on demand.
After this, this paper focuses on IRS-assisted wireless net- The other main issue is the realization of reconfigurability
works. Section III discusses the path loss and signal mod- to achieve full and accurate phase control of the reflected
els, performance metrics, and provides a stochastic analysis or diffracted waves. In the sequel, we first discuss the IRS’s
of IRS-assisted wireless networks, considering the random structural design and inter-cell communications mechanism to
deployment of scattering elements in the radio environment. connect and control all scattering elements. After that, we
Section IV presents the performance optimization of IRS- discuss and compare different phase tuning mechanisms to
assisted wireless networks, typically by a joint optimization of achieve a variety of tunable functions and their applications.
the IRS’s passive beamforming and the transceivers’ transmit We also provide a review on prototypes and experiments in the
control. Section V discusses more emerging use cases of literature to verify the feasibility of using the IRS in practice.
the IRS in wireless communications. Finally, we summarize
some practical challenges and future research directions in
Section VI, and then conclude the paper in Section VII. The A. The IRS’s Structure and Control Mechanism
main topics covered in this paper are shown in Fig. 2. A list 1) IRS Controller and Tunable Chips: In general, the IRS’s
of abbreviations used in this survey is given in Table II. reconfiguration of its EM behaviors is achieved by a joint

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
phase control of individual scattering elements. This implies an
integration of tunable chips within the structure of the metasur-
face, where each tunable chip interacts locally with a scattering
element and communicates to a central controller, e.g., [40],
[41]. Hence, it allows a software-defined implementation of the
control mechanism [42]. For example, the IRS controller can
be implemented in a field-programmable gate array (FPGA)
and the tunable chips are typical PIN diodes [41]. As illustrat-
ed in Fig. 3, the embedded IRS controller can communicate
and receive reconfiguration request from external devices, and
then optimize and distribute its phase control decisions to all
tunable chips. Upon receiving the control information, each
tunable chip changes its state and allows the corresponding
scattering element to reconfigure its behavior. The IRS can
be also equipped with embedded sensors with the capability
of sensing the environment [10]. Such sensing information
can be used by the IRS controller to automatically update
its configuration and thus maintain consistent EM behaviors
under dynamic environmental conditions.
The tunable chips can be PIN diodes with ON and OFF
states. This allows the change of input impedance to match
or mismatch with the free space impedance [40]. The authors
in [43] design and demonstrate through experiments a binary-
state tunable chip based on the hybridized resonator controlled
by a PIN diode as the unit cell of the metasurface, operating
with the resonance frequency around 2.466 GHz. The tunable
chips can also be varactor diodes, which can be adjusted in a
continuous way given different voltage bias, e.g., [44], [45].
The integrated circuits (ICs) with continuously tunable load
impedance are designed in [46] to control the phase shifts of
scattering elements. Both tunable resistance and capacitance
in the ICs can be controlled by imposing upon a gate voltage.
Hence, wave manipulation can be realized by optimizing
the biasing voltages. As an example, the authors in [40]
demonstrate the perfect absorption at 5 GHz for different
polarizations and incidence angles.
The authors in [42] and [47] propose to integrate a network
of tiny controllers within the metasurface and wirelessly
connect it to an external device. Each controller is capable
of interpreting external instructions and tuning its varactor
to achieve a desired impedance configuration. The change of
connectivity at different locations of the controller network
can realize the reconfiguration of the IRS’s physical structure,
resulting in multiple tunable functionalities. In contrast, the
authors in [48] design the IRS by connecting the scattering
elements to a smaller group of controller chips. Each controller
chip serves four metallic scattering elements. The controller
chips can adjust the EM behaviors of the metasurface by at-
tributing additional local resistance and reactance on demand.
Similar to [48], the metasurface designed in [49] is composed
of reconfigurable metamaterial strips arranged in a grid. A
set of four strips is controlled by an intra-tile controller (i.e.,
tunable chip). All intra-tile controllers are interconnected to
constitute the intra-tile network, which can receive external
configuration instructions from gateway controllers.
2) Inter-cell Communications: The IRS’s reconfigurability
depends on the inter-cell communications among the tunable
chips jointly controlling the scattering elements of the meta-
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
Tunable chip or PIN switch
Passive scattering element PCB substrate
x1
z
x0
x y1 y0
z0
y
Embedded controller
Inter-cell communications
Fig. 3: The IRS is made of a reconfigurable metasurface
composed of a large array of passive scattering elements.
surface to exhibit desirable tunable functions. Communications
among the underlying chip controllers can be either wired
or wireless [48]. Wired communication can be preferable as
it can be integrated with the controllers within the same
chip. Wireless inter-cell communication becomes a compelling
alternative in either large-scale or dense metasurfaces. The
design of inter-communications protocols has to comply with
stringent energy, latency, and robustness requirements [50].
The authors in [48] propose two approaches for wireless
inter-cell communications. The first approach exploits the
metasurface structure, while the second approach employs
a dedicated communication channel beneath the metasurface
structure. In the first approach, the communication channel is
the space between the scattering elements and the substrate,
which acts as a waveguide for signal propagation. The second
approach is achieved by adding an extra metallic plate be-
low the chip. The dedicated communication channel provides
an obstacle-free waveguide for information communication
between tunable chips. To ensure reliable communications
among controllers embedded in the metasurface structure,
the authors in [50] adopt the traditional Network-on-Chip
(NoC) methodologies and develop two fault adaptive routing
algorithms, which can bypass the faulty links by using a
properly designed fault-tolerant routing metric.
The authors in [51] develop embedded ICs within the
metasurface structure, which uses mini-routers to move con-
trol information among an arbitrarily-large size of scattering
elements. All ICs are connected to a shared gateway, which
can receive configuration instructions from external wireless
devices. Each IC can configure the load impedance at different
locations of the metasurface. The authors design a handshaking
mechanism to coordinate the information exchange among
different metasurface cells. Such a design can achieve salient
benefits, including delay insensitivity, low EM emissions,
and low power consumption. Medium access control (MAC)
strategies are investigated in [52] to share the wireless medium
efficiently among metasurface cells. The analysis of physical
constraints, performance objectives, and traffic characteristics
of on-chip communications shreds some insights on the MAC
protocol designs for a large number of metasurface cells.
3) Phase Tuning Mechanism: The IRS’s reconfigurability
depends on manipulating the phase of individual scattering
elements. When the external or ambient stimulus changes, the
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
z Tunable chip or PIN switch
x Incident signal
Scattering element
dz
y dy x0
y0
z0 g0
PCB substrate
dx
Fig. 4: Schematic of the unit cell for IRS, e.g., [46]. The basic
topology and dimension parameters (dx , dy , dz , x0 , y0 , z0 , g0 )
can be selected for operations in different frequency ranges.
A tunable chip (e.g., diode and varactor) is incorporated to
provide a variable impedance in a continuous way.
physical parameters of the scattering elements and substrate
will be tuned accordingly. Typical stimulus includes electric,
magnetic, light, and thermal stimulus, which can tune the main
body of a metasurface and thus provide a global control over
its EM properties, e.g., absorption level, resonance frequency,
and polarization of waves. Individual phase control of each
scattering element is also possible, by applying the stimuli
locally to each scattering element. This method is expected to
achieve more sophisticated wave manipulations, such as beam
steering, focusing, imaging, and holography [40].
The most straightforward way to achieve local tuning is
by changing the physical dimension of the scattering element,
resulting in the change of resonant frequency and hence the
phase shift, as illustrated in Fig. 4. The authors in [53]
implement the full controllable phase shifts by combining
two different scattering elements. The overall wave reflections
can be minimized by optimizing the array pattern. More
prevailing tuning approaches are based on the electrically
controlled binary-phase tuning or continuous reactance tuning
mechanism, e.g., by using diodes or varactors, respectively,
e.g., [44]–[46]. The authors in [54] use the electronically
controllable liquid crystal for real-time wave manipulation.
Each metasurface cell is loaded with a thin layer of the
liquid crystal. By controlling the voltage bias on each cell,
the effective dielectric constant will change and therefore
lead to desirable phase shifts at various locations of the
metasurface. Full-wave simulation results verify that effective
beam steering can be achieved in real time. The authors
in [55] experimentally verify that the use of an acoustic cell
architecture can provide enough degrees of freedom to control
a refractive metasurface. In particular, a normal incident wave
can be redirected over 60◦ with the efficiency up to 90%.
B. Typical Tunable Functions
The IRS can support a wide range of tunable functions, such
as perfect absorption, anomalous reflection, beam shaping, and
steering [33], [41], [56], as illustrated in Fig. 5. Moreover, it
is capable of sensing and communicating with the external
devices [10]. This allows it to be integrated with wireless
communication systems and utilized in a variety of wireless
applications, e.g., [11], [12], [57].
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
Signal reflection
Metasurface
Beam focusing Beam steering
Signal reflections
in space
Beam splitting Exotic radiation pattern
Fig. 5: Illustrations of typical tunable functions.
1) Perfect Absorption: The phase shifts of a metasurface
can be designed to ensure minimal reflected and/or refracted
power of the impinging waves. Such a metasurface can be used
to design an ultra-thin invisibility skin cloak, typically used for
visible light [37], [58]. In particular, absorbing metasurface
can be used as carpet cloaking of a scatterer on a flat ground
plane [59]. This is achieved by ensuring the same reflection
angle as the incident angle everywhere on the metasurface.
The authors in [53] propose a single-layer terahertz (THz)
metasurface to produce ultra-low reflections across a broad-
frequency spectrum and wide incidence angles. The authors
in [45] show that perfect absorption can be achieved at an
operating frequency in the range of 4-6 GHz by applying a
change in the reverse biasing voltage of the varactor. In [33],
the authors introduce the design of broadband absorbers of
light by using unit cells featuring with multiple adjacent res-
onances. A tunable metasurface absorber is presented in [60]
by using an optically-programmable capacitor as the tunable
chip of each cell. The designed metasurface can operate
at 5.5 GHz and achieve a tuning bandwidth of 150 MHz.
The authors in [46] realize a reconfigurable metasurface that
exhibits perfect absorption at 5 GHz with different incidence
angles, by changing the capacitance of each cell.
2) Anomalous Reflection: Abnormal reflection can be ob-
served when light beams impinge upon optical metasur-
faces [61]–[63]. The authors in [61] realize anomalous re-
flection of light beams in optically thin arrays of metallic
antennas on silicon. The authors in [62] study the optical
properties of different metasurfaces and present a metasurface
design that can reflect two orthogonal polarization states in
a broad wavelength range. The authors in [63] design optical
metasurfaces to facilitate tuning of the reflection phase and
polarization. The experiment demonstrates the feasibility of
tuning the reflection phase over π by controlling the reso-
nant properties of the antennas. An acoustic metasurface is
constructed in [64], which can tailor the reflected waves with
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
discrete phase shifts covering the full 2π span. The authors
in [65] show that perfect phase control can be realized by
completely eliminating the metasurface’s parasitic reflections
into unwanted directions. The resulting power efficiency of
reflection can be over 90% at 8 GHz. A lossless metasurface
is proposed in [66], where the incident wave can be perfectly
received and converted into a surface wave along the surface,
before it is radiated into space without power loss from a
different location from the receiving point.
3) Wave Manipulation: Wave manipulation/modulation can
create multiple reflections in different directions simultaneous-
ly based on perfect phase control of the metasurface. That is,
the power of the reflected waves can be temporal-spatially dis-
tributed to create an exotic power radiation pattern, as shown in
Fig. 5, which can be exploited to carry information by spatial
modulation [67]. Using PIN diode as binary-state control of
each scattering element, the authors in [41] optimize the binary
coding matrix and thus create different wave manipulations of
a large metasurface, including anomalous reflection, diffusion,
beam steering and beamforming. The real-time switch among
these radiation patterns can be achieved by an FPGA-based
controller. A genetic algorithm is employed in [68] for arbi-
trary wave modulations to create desirable radiation patterns
according to application requirements. The effectiveness of the
genetic algorithm is experimentally verified at an operating
frequency of 10 GHz, showing that the accuracy of wave
modulation increases with the size of scattering elements.
Different from the aforementioned spatial wave modulations,
the authors in [69] verify the possibility of simultaneous
wave manipulations in both space and frequency domains. The
perfect control of the reflection angles and power distribution
can be achieved by using a space-time modulated digital
coding metasurface.
4) Analog Computing: It is also possible for the IRS
to perform more complicated mathematical operations (such
as spatial differentiation, integration, convolution, and even
neural network training) as the impinging wave propagates
through the scattering elements. This is referred to as wave-
based analog computing, achieving a higher energy efficiency
compared to conventional digital signal processing paradigms.
The authors in [70] introduce the concept of metamaterial
analog computing that uses optical metasurfaces to perform
mathematical operations in the spatial Fourier domain. This
offers the possibility of miniaturized, potentially integrable,
wave-based computing systems. Analog computing of acoustic
metasurfaces is also demonstrated in [71] by using thin planar
metamaterials to perform mathematical operation in spatial do-
main. This is promising for various applications including high
throughput image processing, ultra-fast equation solving, and
real-time signal processing. The authors in [72] experimentally
demonstrate that the off-the-shelf wireless infrastructure in
combination with a tunable binary-phase metasurface can
perform analog computation with Wi-Fi signals.
Some other exotic tunable functions also appear in the
literature. The authors in [73] use the metasurface directly
as advanced sensing devices for diagnostic applications, e.g.,
cancer detection, biological tissue characterization and chem-
ical analysis, based on the interactions of EM waves with
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
the metasurface. The authors in [74] show the possibility of
storage and retrieval of EM waves by introducing varactor
diodes to manipulate the metasurface’s structure. The authors
in [75] demonstrate that the slowdown, storage and retrieval of
multi-mode EM waves can be achieved through active manipu-
lation of a control field, which shows the possibility for multi-
mode memory of EM waves and its practical applications in
information processing.
C. Experiments and Prototypes
Prototypes of reconfigurable metasurfaces have been devel-
oped to verify the feasibility of different tunable function-
s. Table III summarizes recent experiments and prototypes
developed in the literature for the verification. The authors
in [41] implement a metasurface containing 1600 individually
controllable cells to demonstrate the feasibility of dynamic
wave manipulations. Each cell of the metasurface is integrated
with one PIN diode that can switch between two states. The
authors in [55] propose and experimentally verify the use of
an acoustic cell architecture to provide enough degrees of
freedom for full phase control. Three refractive metasurfaces
are designed to redirect a normally incident plane wave by 60◦ ,
70◦ , and 80◦ , respectively, with the efficiency up to 90%. The
authors in [82] design a graphene-integrated metasurface to
induce a tunable phase change to the incident wave, which can
be controlled at an ultra-fast speed. The designed prototype
is shown to change the reflection phase up to 55 degrees.
The authors in [77] develop a large-scale THz all-dielectric
metamaterials with the outer dimension 900 cm × 900 cm
by using the template-assisted fabrication method. Using such
metamaterials, the authors implement a broadband reflector
with a bandwidth of 0.15 THz and demonstrate its reflection
up to 95%, which implies a wide variety of applications in
low-loss and high-efficiency THz devices.
The development of reconfigurable metasurfaces shares
a similar idea with the classical concept of reconfigurable
reflectarrays, in which the resulting radiation pattern of the
signal transmitter is altered as desired by changing the current
distribution [12]. Comparing to reconfigurable reflectarrays,
IRS is featured with the real-time control and reconfigurability
of the radio environment [3]. A reconfigurable reflectarray
antenna is implemented in [78], consisting of a group of
244 radiating elements phase controlled by PIN diodes. The
antenna is designed to operate in the band from 10.10 GHz to
10.70 GHz, capable of switching the beam between −5◦ and
5◦ . Different from the PIN diodes in [78], the authors in [79]
implement a 6 × 6 transmitarray controlled by varactor diodes
to verify its beamforming capability. Experiments demonstrate
its capability of beam scanning over a 100 × 100-degree
window at the operating frequency of 5 GHz. To reduce the
size of transmitarray, the authors in [83] design and verify a
compact reconfigurable antenna for wireless communications,
which is capable of generating four different radiation patterns
at the operating frequency of 2.45 GHz.
VISORSURF is an interdisciplinary program funded by
Horizon 2020 FET-OPEN [49], [84]. Its objective is to develop
a hardware platform for reconfigurable metasurfaces, namely
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
TABLE III: Summary of Experiments and Prototypes of Reconfigurable Metasurfaces
Reference Controller Dimension
[41] FPGA 40 × 40 array
[55] – 7 cells
[57] FPGA 16 × 16
[64] – 8 cells
[68] – 40 × 40 array
[69] FPGA 8 × 8 array
[76] DC voltage 22 × 22 array
source
[77] – 900 × 900 cm
[78] – 244 cells
[79] – 6 × 6 array
[80] Bias network 21 × 21 cm
[81] MEGA 2560 14 × 16 array
micro-controller
hypersurface, featured with programmable EM behavior by
controlling a network of switches. A prototype of the re-
configurable metasurface is implemented in [76] to achieve
multi-functional control of EM waves, e.g., beam splitting,
deflection, and abnormal reflection at microwave frequencies.
This is achieved by controlling the PIN switches and varactor
diodes associated with each scattering element. The real-time
switch between different EM functionalities is controlled by a
computer-controlled voltage source. Realizing wave manipula-
tions in both space and frequency domains, the authors in [69]
design a space-time modulated digital coding metasurface
to control the propagation direction and power distribution
simultaneously in the frequency range of 8-10 GHz. The
experimental results demonstrate a good performance for beam
steering, beam shaping, and scattering-signature control. The
authors in [57] develop an IRS prototype with 256 reflecting
elements and verify its performance gains in wireless commu-
nications, i.e., a 21.7 dBi antenna gain can be obtained at 2.3
GHz and a 19.1 dBi antenna gain can be achieved at 28.5 GHz
mmWave frequency. The power consumption by using the IRS
can also be reduced significantly. Reference [81] implements a
60 GHz reconfigurable reflect-arrays to assist IEEE 802.11ad-
based mmWave communications, creating additional reflection
links when there exist no line-of-sight (LOS) links.
In the following, we focus on the applications of the IRS in
wireless communications. We first introduce the recent works
discussing the novel concept of a smart/programmable wireless
environment, which is envisioned to change the wireless
networking paradigm by using the IRS in wireless communi-
Communications Surveys & Tutorials
Phase Control Frequency range Realized Tunable Functions
Binary PIN diodes 9-12 GHz Wave manipulation including anomalous reflection,
diffusion, shaped scattering
– 3.0 GHz Redirecting a normal incident wave to 60◦ , 70◦ , and
80◦ , with efficiency over 90%
2-bit IRS element 2.3 GHz, 28.5 21.7 dBi antenna gain obtained at 2.3 GHz, 19.1 dBi
GHz antenna gain achieved at 28.5 GHz mmWave
Traverse length Acoustic Discrete phase shifts covering the full 2π span with
steps of π/4
Cell layout 8.7-11.3 GHz Generating different radiation patterns. Even distri-
bution of reflection phases from 0 to 2π.
Binary PIN diodes 8-10 GHz Simultaneous wave manipulations in both space and
frequency domains
PIN switches and var- 11.5-13.5 GHz Dynamical beam deflection, splitting, and polariza-
actor diodes tion. 180◦ reflection phase difference
Geometric parameters 0.6-0.9 THz Broadband reflector with a bandwidth of 0.15 THz
and efficiency up to 95%
Binary PIN diodes 10.10-10.70 GHz Beam switching between −5◦ and 5◦
Varactor diodes 5 GHz Beam scanning over a 100◦ × 100◦ window
Binary PIN diodes 2.75-4.0 GHz Switching between total reflection and absorption
Binary relay switches 60 GHz (IEEE Robust link between transceivers can be established
802.11ad) by steering the incident signal to the desired receiver
radio environment, such as the walls and ceilings, or even
carrying IRS in aerial platforms, such as floating balloons and
UAVs [85]. This implies that the IRS’s scattering elements
are distributed in nature relating to the spatial distribution
of the environmental objects. Therefore, the modeling and
performance analysis of IRS-assisted wireless networks ne-
cessitate analytical models that take into account i) the spatial
locations of IRS units, ii) the IRS’s EM properties, and iii) the
wave manipulations adopted by the coexisting IRS units in the
environment. In this part, we first review two typical channel
models for IRS-assisted wireless communications, and then
present the analytical studies on performance limit of IRS-
assisted wireless networks, including performance bounds and
asymptotic behaviors that are hard to obtain from simulations.
Such a theoretical performance analysis can provide design
insights on the deployment and configuration of IRS-assisted
wireless networks without the need for extensive simulations.
A. IRS-enhanced Path Loss Models
The potential performance gain of using IRS in wireless
networks can be verified by developing a path loss model
to analyze the received signal power and SNR performance.
The authors in [12] derive a simplified path loss model based
on the conventional two-way channel model for wireless
communications. As shown in Fig. 6, in addition to the direct
path, each IRS’s reflecting element provides a second path
from the transmitter to the receiver, constituting a two-way
signal propagation model. Combining the signals from all
paths, the received signal power can be formulated as follows:
cations. As such, we review the system modeling, performance 2
 2  N
analysis, and optimizations of IRS-assisted wireless networks. λ  1 X Rn e−j∆φn 

Pr = Pt  +  , (1)
4π ` d + d2,n 
n=1 1,n
III. PERFORMANCE ANALYSIS OF IRS-ASSISTED
where N denotes the number of IRS’s reflecting elements
WIRELESS NETWORKS
and ` denotes the distance of direct path, which can be
The vision of the smart radio environment can be realized approximated by the distance d between the transceivers. d1,n
by coating IRS to the facades of physical objects in the and d2,n denote the distances of two segments in the reflected

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Distance 𝑑
ℎ chamber. In particular, given the transmit power and the
𝑑 , 𝑑 ,
Transmitter IRS
IRS controller
Fig. 6: Two-way channel model for IRS-assisted wireless
communications.
path via the n-th reflecting element. The summation term in (1)
denotes the signal reflections via different paths. The phase
difference ∆φn is determined by the distances of the direct
path and the reflected path via the n-th reflecting element.
Rn denotes the reflection coefficient depending on the EM
properties of the reflecting object, which is conventionally
uncontrollable without using IRS.
For each reflecting element on the IRS, we can proactively
control its phase shift Rn = ej∆φn such that the reflected
signal can be coherently aligned with the direct path. Normal-
ly, we can assume d ≈ ` ≈ d1,n + d2,n , which leads to the
following approximation of the received signal power in (1).

λ
2
2
Pr ∝ (N + 1) Pt . (2)
4πd
If there is no direct link or the size N of reflecting elements
becomes large, the
above path loss can be simply rewritten
λ 2
as Pr ≈ N 2 Pt 4πd . Compared to the free-space path loss
model, the IRS-assisted path loss in (2) introduces additional
gain to the received signal power, i.e., the received power is
proportional to N 2 . For an IRS with 100 reflecting elements,
the power gain amounts to a significant 40 dB. The authors
in [12] also derive the SNR performance in a single-antenna
point-to-point system under flat fading channel conditions. For
a large number of scattering elements with optimal phase
configuration, the compound channel from the transmitter
to the receiver can be viewed as a Gaussian distributed
random variable with the mean and variance proportional to
N . This implies that the average received SNR will increases
proportionally to N 2 . The distribution of SNR can also be
characterized and used to derive the symbol error probability
of IRS-assisted wireless communications. The upper bound
of the symbol error probability is shown to have a waterfall
region when SNR is low while fall into a slowly-decaying
region for higher SNR. In both regions, the error probabilities
can be decreased significantly by using a larger IRS.
The simplified path loss model in (2) fosters basic un-
derstanding of IRS’s superiority in wireless communication-
s, however omits complex physical details during the IRS-
assisted signal propagation, e.g., the size and dimension of
the reflecting elements, the angles of incident and reflected
signals, and the anisotropic radiation pattern of antennas at
the transceivers. Based on a detailed study on the IRS’s
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
Communications Surveys & Tutorials
10
physics and EM nature, the authors in [86] propose a more
reliable free-space path loss model for IRS-assisted wireless
communications. Extensive simulation results validate that the
proposed channel model matches well with the experimen-
tal measurement results conducted in a microwave anechoic
Receiver ℎ
positions of transceivers, the optimal received signal power
in the far-field can be characterized as follows:
 2  2 
2 λ ρ F (θt , φt )F (θr , φr )
Pr ∝ Gc N Pt , (3)
4πd1 d2 4π
where d1 , d2 denote the distance from the transmitter to
the center point of the IRS, and from the center point to
the receiver, respectively. The function F (·, ·) determines
the antennas’ power radiation pattern at the transceivers. In
particular, F (θt , φt ) returns the normalized power level when
the transmitter has the elevation and azimuth angles given by
(θt , φt ), with respect to the IRS’s center point. The constant ρ
denotes that all reflecting elements use the same magnitude of
reflection coefficients. From the analytical results in [86], the
coefficient Gc depends on the antenna gains at the transceivers,
the dimension and power radiation pattern of the unit reflecting
element. Similar to (2), the path loss model in (3) implies
that in both cases the received signal power is proportional
to N 2 . A different observation is that the far-field received
signal power in (3) depends on the product-distance, i.e., it
is proportional to 1/(d1 d2 )2 , instead of the sum-distance
1/(d1 + d2 )2 in (2). However, the analytical results in [86]
reveal that the sum-distance rule can hold for the received
signal power in the radiative near-field. A similar result has
been revealed in [87].
The signal model of IRS-assisted wireless communications
can be established based on the two-way channel model. Con-
sidering a simple case with single antenna at the transceivers,
let f1,n and f2,n denote the complex channels before and
after the reflection point, g be the complex channel for the
direct link, the IRS-assisted channel from the transmitter to
the receiver is thus given by
N
X
g̃ = g + ρn exp (jφn )f2,n f1,n = g + f1H Θf2 ,
n=1
where ρn and φn represent the magnitude and complex phase
of the reflection coefficient at each reflecting element, which
can be tuned properly such that the reflected signal is aligned
with the direct channel g. f1 and f2 denote the channel
vectors of f1,n and f2,n . The matrix Θ has the diagonal vec-
tor given by [ρ1 exp (jφ1 ), ρ2 exp (jφ2 ), . . . , ρN exp (jφN )]H ,
which denotes the reflecting coefficients of each reflecting
element. For a multi-antenna case, the IRS-assisted channel
can be characterized in a similar form. Most of the existing
analytical works are built based on the above path loss and
signal models, which are basis for the following performance
analysis and optimal design of IRS-assisted wireless systems.
B. Using IRS as the Signal Reflector
The performance metrics of interest mainly include the
probabilistic metrics to characterize the uncertainty in wireless
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
transmissions of individual transceivers, as well as the ergodic
metrics to characterize the averaged network performance,
considering the randomness in network topology, the dis-
tribution of reflectors, channel conditions, and interference
variations, etc. Typical performance metrics proposed for IRS-
assisted wireless systems include the following aspects.
• Reflection probability: The probability that an IRS can
reflect the signals from a transmitter to the receiver.
• Coverage or outage probability: The probability that the
received SNR is above or below a target threshold.
• Bit error probability: The probability that the decoded
information bit differs from the transmitted one.
• Ergodic capacity: The expectation of channel capacity
measured by Shannon’s formula.
• Transport capacity: The aggregated data rate that can be
reliably communicated in the entire system.
1) Probabilistic Performance: The reflection probability is
studied in [88] considering a large-scale IRS randomly dis-
tributed by a Boolean model of line segments [89]. The authors
derive the exact probability that a random located IRS can
reflect for a given transceiver according to the generalized laws
of reflection. The analytical results reveal that the reflection
probability of a randomly located reflector is independent of
its length. However, this work assumes that all the IRSs have
a fixed length, which cannot capture the real-world network
environment. Besides, the authors only analyze the reflection
probability, without an evaluation on the improvement of
transmission performance by using the large-scale IRS. The
authors in [81] propose the use of the IRS to improve the LOS
probability for indoor mmWave communications. Compared
to the existing relay-based approaches that create LOS links
for mmWave communications, the reflection-based approach
exempts from self-interference of a full-duplex relay and
compromised throughput of a half-duplex relay.
Considering the downlink transmission from an mmWave-
based access point (AP) to a user with only the reflection links
(i.e., no direct link due to obstacles), the authors in [81] derive
an expression for the outage probability and further minimize
the outage probability by optimizing the IRS’s deployment
position. The authors in [90] study the outage probability of
IRS-assisted systems under Rician fading where the IRS’s
phase shifts only adapt to the LOS components. The outage
performance is firstly analyzed and optimized in the slow
fading scenario for the non-LOS components. It can be shown
that the optimal outage probability decreases with the size of
the IRS when the LOS components are stronger than the non-
LOS ones. Then, the authors characterize the asymptotically
optimal outage probability in the high SNR regime, and show
that it decreases with the powers of the LOS components.
The coverage probability is studied for an IRS-assisted
wireless network in [91]. Different from [81] which studies
point-to-point mmWave communications, the authors in [91]
consider a generalized mmWave downlink cellular network
coexisting with random obstacles and reflectors. A stochastic
geometry method is proposed to analyze the downlink cov-
erage probability under the assumption that the locations of
base stations follow a homogeneous Poisson point process, the
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
11
RF communications
Signal receiver
Interference
Reflections
Reflections
Signal
transmitter
IRS controller
Fig. 7: The IRS is used for enhancing spectrum sharing among
multiple transceivers.
blockages and reflectors are deployed in straight line segments
with uniformly distributed orientation and length. The study
indicates that only the deployment of reflectors with high
density can cause a noticeable improvement in the mmWave
coverage. However, the deployment with low density may not
benefit mmWave coverage as the reflected signals go through
longer distances than the direct links. A limitation of this
work is that only the reflections from the nearest reflector
is considered. In fact, reflections from other nearby reflectors
could also be strong and should not be ignored.
In contrast to above efforts that focus on an analytical study,
reference [92] presents a joint analytical and empirical study
for the bit error probability of spatial modulation based on
reconfigurable antennas, which encodes information bits on
the radiation patterns of a reconfigurable antenna. To evaluate
the impact of the radiation patterns on the error performance,
the authors in [92] introduce an analytical framework to char-
acterize the bit error probability and identify the best radiation
pattern to minimize the average error performance. In [93], the
BER performance is derived in closed form for an IRS-assisted
non-orthogonal multiple access (NOMA) downlink system.
The authors in [94] present a general mathematical framework
to derive the symbol error probability of an IRS-assisted
wireless system, where the IRS acts either as a reflector or
an access point with a simple transceiver architecture. The
numerical results verify that ultra-reliable communications can
be achieved by using the IRS to boost the received SNR.
2) Ergodic Capacity and Data Rate: The authors in [95]
analyze the asymptotic achievable rate in an IRS-assisted
downlink system. A passive beamformer is designed to achieve
the optimal asymptotic performance by tuning the EM proper-
ties of signal waves. To maximize the data rate, a modulation
scheme is designed for the IRS that is interference free
for existing devices. The authors also derive the expected
asymptotic symbol error rate (SER) of the considered system
and propose a protocol for joint user scheduling and power
control. Simulation results verify that the achievable rate in
a practical IRS-assisted system satisfies the asymptotic opti-
mality. Considering an IRS-assisted mmWave MIMO system,
the authors in [96] characterize the achievable data rate from
the BS and to a mobile user, by designing the IRS’s optimal
phase shifts based on limited feedback from the mobile user.
Besides improvement on data rate, simulation results show that
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
the positioning error bound and orientation error bound both
can be reduced by using the IRS with perfect CSI.
From another aspect, the authors in [97] consider achieving
capacity gain of a point-to-point MIMO system by optimizing
the IRS’s phase matrix to improve the rank of channel matrix.
The IRS-assisted rank improvement enriches the propagation
conditions by adding more multi-paths with different spatial
angles. With 100 reflecting elements, the numerical results
show that the capacity gain can be over 100% by carefully
deploying the IRS’s location and optimizing its phase matrix.
The aforementioned works focus on the use of ideal IRS
with infinite phase resolution, and thus the resulting capacity
analysis bears an undetermined gap with the practice. The
authors in [98] derive an approximation of the achievable
data rate and characterize the performance degradation when
a practical IRS is implemented with limited phase shifts.
Considering hardware impairments in the IRS’s reflecting
elements, the authors in [99] model the correlation structure
of the hardware impairments by a function of the distance
between reflecting elements and analyze the degradation of
achievable rate by using a simple receiver structure.
Different from the prior works that focus on link-level
capacity maximization for IRS-aided wireless systems, the
authors in [100] characterize the spatial throughput of a single-
cell MU downlink system assisted by multiple randomly
distributed IRSs. The spatial throughput is averaged over
the distributions of all users and IRSs’ random locations.
The simulation results show that the IRS-assisted system can
achieve higher spatial throughput than that of a full-duplex
relay system. To maximize the spatial throughput, the authors
reveal an interesting throughput-fairness tradeoff, which shows
that it is preferable to have fewer IRSs with more reflecting
elements, however at the cost of more diverse distribution of
user rates. The authors in [101] analyze the capacity region of
a multiple access channel where two users are assisted by IRS
units to send information to an AP. The deployment of IRS
units can be either distributed or centralized. The distributed
IRS units are closer to the users while the centralized IRS is in
the vicinity of the AP. For both deployment schemes, capacity
regions can be characterized and compared analytically. It
can be shown that the centralized scheme outperforms the
distributed scheme under the practical channel setup. Both
works verify that the IRS-assisted system generally achieves
higher capacity performance, which becomes more significant
with asymmetric user rates. Hence, we may envision that the
IRS has the superiority for the fulfillment of diverse user
requirements in future wireless networks.
For multiple transceiver pairs coexisting in one system, the
capacity analysis becomes more involved as it depends on the
multiple access protocols. The authors in [102] propose the
IRS-assisted spectrum sharing scheme for multiple transmitter-
s. As shown in Fig. 7, the idea is to use the IRS for steering the
signal beams from different transmitters to enhance the useful
signals and cancel the interference towards their respective
receivers. This paradigm allows multiple transmitters to simul-
taneously operate on the same spectrum band without causing
interference to each other. Considering an indoor wireless
scenario using the spectrum sharing scheme, the authors derive
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
12
Signal transmitter
IRS controller
Signal receiver
Fig. 8: The IRS is used as the signal receiver to enhance system
capacity.
both upper bound and achievable bound of the transport
capacity under practical deployment constraints. A test-bed
is also developed to study the effect of the system parameters
and validate the practicality of the proposed spectrum sharing
concept. The experimental results demonstrate that the use
of the IRS realizes a significant improvement on spectrum
efficiency for the legacy transceivers. However, the test-bed
only implements two transceivers. The practical performance
of a large number of indoor communication pairs sharing the
same spectrum band is still unknown.
C. Using IRS as the Signal Receiver
Instead of using the IRS as a signal reflector for performance
enhancement of wireless communications, the authors in [103]
examine the performance of using the IRS as the signal
receiver. In particular, the authors consider an uplink scenario
where multiple single-antenna devices distributed in a three-
dimensional (3D) space transmit to a vertical plane metasur-
face (e.g., hung on the wall), as shown in Fig. 8. Assuming
large-size IRS and LOS channel conditions, the normalized
capacity per unit of the IRS with perfect channel estimations
can be characterized into an integral expression. The analytical
expression demonstrates an asymptotic behavior that, given a
constant transmit power, the normalized capacity approaches
half of the transmit power divided by the spatial power spectral
density of noise. The authors in [104] consider a similar MU
MIMO system, where the uplink transmissions to the IRS need
to cancel interference among different users by optimizing
the IRS’s phase shifters. Based on properly design message
passing schemes, the authors in [104] propose decentralized
algorithms for the reflecting element to iteratively update
phase shifters simultaneously. However, both works in [103]
and [104] focus on idealistic models, assuming either perfect
channel information or full control of the phase shifts.
Different from [103], reference [105] considers a more
practical scenario where the IRS has a finite area and the
interfering channels could either be LOS or non-LOS. With
imperfect channel estimations, the authors derive the asymp-
totic capacity and reveal a channel hardening effect, i.e., the
impacts of channel estimation errors and the non-LOS path are
negligible when the number of scattering elements becomes
large. The simulation results also demonstrate that, compared
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
to conventional massive MIMO, a large-scale IRS can achieve
a better reliability in terms of expectation and variance of
capacity. Imperfect channel estimation is taken into account
in [106], where multiple IoT devices are connected to multiple
non-overlapping IRS units in an IRS-assisted uplink system.
The asymptotic sum-rate distribution with Rician fading can
be derived in closed form and used for the analysis of outage
probability. Numerical results verify that the IRS can provide
reliable communications in terms of outage probability. The
authors in [107] investigate the performance of IRS with
hardware impairments which induce a higher effective noise
level. A closed-form expression is derived to characterize the
capacity of the uplink transmission to the IRS. To mitigate the
negative effects of hardware impairments, the analytical and
simulation results suggest a splitting of a large-scale IRS into
an array of smaller IRS units. Different from the previous
works focusing on capacity performance, reference [108]
explores the potential of using an IRS with a large number
of scattering elements for terminal positioning. The analysis
shows that deploying multiple IRSs in a distributed manner is
effective to improve the coverage for terminal positioning.
D. Using IRS as the Signal Transmitter
Besides using IRS as a reflector or a receiver, it is also
possible to use IRS as a source signal transmitter, similar
to wireless backscatter communications that adapt the an-
tenna’s reflection coefficients by load modulation [12], [27].
By controlling the phase shifts of the IRS’s scattering ele-
ments, the outbound waves generated by the IRS can create
different radiation patterns, which can carry information if
these radiation patterns can be sensed and differentiated at the
receiver. This is the main design principle of spatial modula-
tion, which is conventionally realized by using reconfigurable
antennas [109]–[111]. The IRS’s superior reconfigurability and
capability of reshaping the wireless environment make it much
easier for its applications in spatial modulation [12]. Along
this research direction, the authors in [112] design spatial
modulation methods for the IRS, which can be realized by
either creating exotic scattering pattern to enhance end-to-end
transmissions or utilizing the index modulation of multiple
receive antennas to render improved spectrum efficiency. Nu-
merical results demonstrate that IRS-based spatial modulation
methods provide a large capacity with ultra-low BER. The
authors in [12] derive the error probability in closed form
by using IRS to transmit M-ary phase modulated symbols.
It shows that the IRS can transmit information with high
reliability, similar to using it as a reflector.
The authors in [28] use IRS to perform wireless backscatter
communications leveraging its capability of manipulating the
phases of the reflected signals. The authors in [30] similarly
use an IRS as the backscatter transmitter and propose a general
system model for the IRS-based ambient backscatter com-
munications. Considering a large size of reflecting elements,
we expect that the IRS-based backscatter communications
will have higher flexibility than the conventional wireless
backscatter communications, as the IRS is able to generate
more exotic reflection patterns that can be used for information
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
13
communications. This may lead to a higher data rate and
larger transmission distance. In [113], the IRS is used to
simultaneously enhance the legacy RF communications and
send its private data to a receiver by spatial modulation. The
information from both the RF transmitter and the IRS can be
retrieved by a two-step decoding scheme. Similar to [113],
the authors in [29] and [114] use IRS to perform backscatter
communications to deliver its own data and simultaneously
enhance the primary communications. An approximation is
derived in [114] for the probability of the backscatter channel
gain greater than the direct link, which is a useful performance
measure to determine the number of the IRS’s reflecting
elements. The analysis and numerical results show that the
use of IRS can achieve high capacity even without the LOS
link. The authors in [115] integrate the IRS’s backscatter com-
munications into a cognitive radio network (CRN), promoting
spectrum sharing between the secondary and primary users.
The IRS on one hand assists the primary user’s RF commu-
nications, on the other hand backscatters its own information.
The design objective is to maximize the secondary user’s data
rate, subject to a QoS requirement at the primary user, by
jointly optimizing the secondary user’s transmit power and
the IRS’s passive beamforming in the alternating optimization
framework. Simulation results show that the IRS-assisted
CRN is efficient for secondary transmissions, even under a
challenging scenario where the secondary transmitter is much
closer to the primary receiver.
Summary: Table IV summarizes and compares the existing
literature reviewed in Section III. The majority of these
research works studies the performance of a standalone IRS
system, i.e., either point-to-point communications or multiple
communication pairs. References [88] and [91] analyze the
performance of deploying the IRS reflectors in large-scale
systems, however, under unrealistic assumptions such as the
fixed-length model for random environmental reflectors [88]
and LOS links for all the reflectors [91]. Besides, the channel
information is mostly assumed to be known for the IRS’s
phase control, e.g., [92]–[96]. Hence, there is a need for more
realistic models to analyze the performance of using the IRS
in practical large-scale communications systems. Moreover,
the above-mentioned literature hardly takes the user mobility
into account, neither for indoor nor outdoor scenarios. User
mobility introduces not only handoffs among different IRS
units but also a spatial correlation in user distribution that
may cause non-negligible impacts on the system performance.
Hence, it becomes a critical research direction in the future
to incorporate different mobility models into the performance
analysis of IRS-assisted wireless systems. Additionally, the
performance metrics currently under investigation are also
limited in the literature. The theoretical performance of IRS-
assisted wireless systems can be more thoroughly understood
from the potential aspects as follows:
• Pairwise error probability which measures the probability
that the decoded signal is a certain symbol given the
transmitted signal.
• Average area spectral efficiency which is the sum of the
capacity of all the communication channels normalized
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

14

TABLE IV: Summary of Existing Literature in Performance Analysis of IRS Systems

Reference Performance Metrics System Model Characterization

[88] Reflection prob. Point-to-point communications in the presence of random metasurfaces Exact
[81] Outage prob. Point-to-point mmWave communications with only reflection links Exact
[91] Coverage prob. Downlink mmWave communications in the presence of LOS blockages and reflectors Approximation
[92] Bit error prob. Point-to-point MIMO communications Exact
[93] Bit error rate MU MISO NOMA downlink Exact
[94] Symbol error prob. Point-to-Point SISO with Rayleigh fading channels Upper bound
[95] Achievable rate MU MISO downlink in the TDMA protocol with practical limitation of IRS Asymptotic
[96] Achievable rate mmWave MIMO downlink systems with limited feedback from the mobile user Asymptotic
[97] Achieving rate Point-to-point MIMO system with rank improvement Approximate
[98] Achievable rate IRS-assisted uplink system with finite resolution IRS Approximate
[100] Spatial throughput Single-cell MU downlink system with multiple randomly distributed IRSs Approximate
[101] Capacity region A multiple access channel for two users sending information to an AP Upper bound
[102] Transport capacity Spectrum sharing among multiple transceivers through a reconfigurable metasurface Upper bound
[103] Normalized capacity Multiple devices in 3D space transmit to a vertical metasurface with LOS interference Asymptotic
[106] Capacity, outage prob. MU uplink transmission to the IRS with imperfect channel estimation Asymptotic
[105] Ergodic capacity MU uplink transmission to a vertical metasurface with LOS and non-LOS interferences Asymptotic
[107] Ergodic capacity A single-antenna user transmits to a vertical 2D metasurface with hardware impairments Upper bound
[108] Positioning coverage Positioning for a single-antenna device located in front of an IRS Lower bound

over spectral bandwidth and spatial area.
• Energy efficiency which measures the capacity normal-
ized over the energy consumption of IRS systems.
• Handoff rate which is the frequency of occurrences that
a user handoffs to another IRS.
The stochastic performance analysis for IRS-assisted
wireless systems mainly investigates the potential of the
information-theoretic performance gain. For a specific net-
work design problem, the maximization of performance gain
requires a joint optimization of the active transceivers and the
IRS’s passive scattering elements. The joint phase control of
scattering elements can be regarded as passive beamforming,
which is closely coupled with the control variables of the ac-
tive transceivers. This not only makes the performance analysis
of IRS-assisted wireless systems more complicated, due to
randomness and ubiquity of the scattering elements in the radio
environment, but also results in new optimization problems
that require novel solutions to account for the interactions
between active and passive devices.
IV. A PPLICATION AND O PTIMIZATION OF IRS-A SSISTED
W IRELESS N ETWORKS
By smartly adjusting the phase shifts of all scattering
elements, as illustrated in Fig. 1, the reflected signals can
be combined coherently at the intended receiver to improve
the received strength or combined destructively at the non-
intended receiver to mitigate interference. This can be verified
by the experimental demonstration and channel measurements
in [86], which pave the way for further theoretical studies and
system optimization. In the sequel, we review the main opti-
mization formulations and solutions proposed for IRS-assisted
wireless systems. The typical design objectives include SNR
or rate maximization, transmit power minimization, EE/SE
performance maximization, and physical layer security issues.
A. SNR or Capacity Maximization
1) IRS-assisted Point-to-Point Communications: Consid-
ering a point-to-point communication scenario, the authors
in [116] focus on an IRS-assisted multiple-input single-output
(MISO) system, where one IRS with N passive scattering
elements is deployed to assist the downlink information
transmission. A joint beamforming problem is formulated to
maximize the received signal power at the user, by jointly
optimizing the AP’s transmit beamforming and the continuous
phase shift of each scattering element. Semidefinite relaxation
(SDR) is firstly proposed to obtain an approximate solution as
a performance upper bound. Then, the alternating optimization
is employed to update the active and passive beamforming
strategies iteratively. Given the fixed passive beamforming,
the AP’s optimal beamforming is easily obtained by the
maximum-ratio transmission strategy. Given the AP’s beam-
forming, the IRS’s optimal passive beamforming can be simply
aligned with the direct channel to enhance the received signal
power. Compared to the non-IRS-assisted MISO system, the
SNR of the IRS-assisted MISO system can be improved by
around 10 dB using an IRS with 100 scattering elements.
Another important finding in [116] is that the SNR at the
receiver increases in the order of N 2 . This power scaling law
is further studied in [117] and compared with the massive
MIMO system. Analytical results show that a large number of
reflecting elements are required to obtain the SNR comparable
to massive MIMO systems. The authors in [118] consider
a similar IRS-assisted MISO downlink system. The joint
optimization of the AP’s transmit beamforming and the IRS’s
phase shifts is solved by fixed point iteration and manifold
optimization techniques, respectively, which are shown to be
effective in tackling the IRS’s unit modulus constraints. These
two algorithms not only achieve a higher data rate but also
have a reduced computational complexity.
The above IRS-assisted MISO downlink model and the
heuristic alternating optimization in [116] provide a general
framework for the optimized design of IRS-assisted systems,
and thus can be extended to an upsurge of different network
scenarios. The IRS-enhanced MISO OFDM downlink system
is studied in [119]. To maximize the downlink achievable rate,
a joint optimization of the BS’s transmit power allocation

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
and the IRS’ s passive beamforming follows the alternating
optimization framework. The MISO cognitive radio system
is studied in [120], where multiple IRSs are employed to
maximize the data rate of the secondary receiver, subject to the
interference constraints at the primary receivers. The authors
in [121] also use multiple IRSs to assist mmWave MISO com-
munications. Through joint active and passive beamforming
optimization, the IRS can provide enhanced paths of reflection
for mmWave signals, and thus can maximize the received
signal power and extend the network coverage. Both of the
phase shifts and transmit beamforming are derived optimally
in closed forms for the single-IRS case. This is achieved
by exploiting the characteristics of mmWave channels, i.e.,
assuming a rank-one channel matrix between the BS and
the IRS. The analysis and simulation results reveal that the
received signal power increases quadratically with the number
of reflecting elements, which verifies the power scaling law
revealed in [116].
Different from the MISO systems in [118], [119], [121],
the authors in [122] characterize the capacity limit of an
IRS-assisted MIMO system, by jointly optimizing the IRS’s
reflection coefficients and the MIMO transmit covariance ma-
trix. The capacity maximization for broadband transmissions
is considered in frequency-selective fading channels, where
transmit covariance matrices can be optimized for different
OFDM sub-carriers. Based on convex relaxation, the alter-
nating optimization algorithm used in [116] is modified to
find a high-quality sub-optimal solution. Numerical results
show that the IRS-assisted MIMO system achieves substantial
capacity improvement compared to traditional MIMO systems
without IRS, e.g., the improvement is over 45% in the high
SNR regime when using an IRS with 80 scattering elements.
The point-to-point mmWave MIMO OFDM system is studied
in [123]. The hybrid MIMO beamforming matrices and the
IRS’s phase shift matrix are separately optimized to achieve
better BER and spectrum efficiency performance than the
conventional approaches. Considering a similar IRS-assisted
MIMO system, the authors in [124] derive an upper bound for
the ergodic capacity with Rician fading channel. To maximize
the capacity upper bound, the IRS’s phase shift matrix is
optimized by SDR and Gaussian randomization methods.
In [125], the data rate maximization of an IRS-assisted MIMO
system is formulated into a mixed integer problem. Based on
the alternating optimization framework, the ADMM algorithm
is leveraged to find the phase shifts of individual scattering
elements, and then the active beamforming is obtained by
classic singular value decomposition and water-filling solu-
tions. Focusing on a two-way full-duplex IRS-assisted MIMO
system, the authors in [126] maximize the sum rate by a
joint optimization of the precoders at two transmitters and
the IRS’s phase shift matrix. The iterative solution follows a
similar alternating optimization framework as that in [116],
[122]. Comparing to conventional relay-assisted full-duplex
MIMO system, the performance of the IRS-assisted system
is comparable to that of a relay with transmit power at around
−40 dBm to −35 dBm. We note that all aforementioned
research works in [118], [119], [121]–[124], [126] rely on the
ideally designed IRS with infinite phase resolution, i.e., the
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
15
phase shift of each scattering element can be fully controllable.
However, this is difficult for practical implementation and
also complicated for designing exact phase control algorithms.
Moreover, full CSI is generally required for the IRS controller
to make perfect phase control. This implies that the overhead
of information exchange can be prohibitively high, especially
for self-sustainable IRS via wireless energy harvesting.
Ideally, the reflection amplitude is constant and assumed
to be independent of the phase shift. Based on experimental
results, the authors in [127] notice the non-ideality in control-
ling the IRS’s phase shifts, and thus propose a practical phase
shift model that captures the nonlinear dependence between
phase and amplitude. Employing this new model in an IRS-
assisted MISO system, a joint beamforming design problem is
formulated to maximize the achievable rate in MISO downlink
transmissions, which can be solved in the alternating opti-
mization framework. The capacity degradation is evaluated
in [128] by considering different levels of quantization in
the IRS’s discrete phase control. Based on statistical CSI, a
tight approximation of the ergodic capacity is derived for an
IRS-assisted MISO system, and maximized by optimizing the
IRS’s phase shift matrix. Numerical results show that a 2-bit
phase quantizer is sufficient to ensure capacity degradation
of no more than 1 bit/s/Hz. This greatly simplifies the prac-
tical implementation, design costs, and applications of IRS
in wireless communications. In [120], the imperfect CSI is
modeled by a norm-based uncertainty set and a worst-case
robust optimization is formulated to optimize the joint active
and passive beamforming strategy. Considering the channel
estimation errors and training overhead, the authors in [129]
formulate an optimization problem to maximize the achievable
data rate by designing the IRS’s discrete phase tuning strategy.
A low-complexity successive refinement algorithm is devised
to achieve a high-quality sub-optimal solution with proper
algorithm initialization. Similarly, considering discrete phase
shift and unknown CSI in [130], a tight lower bound is derived
for the user’s asymptotic rate in IRS-assisted MISO downlink
communications. Numerical results reveal the discrete phase
design with moderate to high phase resolutions can asymptot-
ically approach that of the optimal continuous phase control
with perfect CSI.
To reduce the overhead in channel estimation, the authors
in [131] propose an element grouping method to exploit
the channel spatial correlation in an IRS-assisted OFDM
system. By estimating the combined channel of each group,
the training overhead can be substantially reduced. The al-
ternating optimization method is also used to maximize the
achievable rate by optimizing the BS’s power allocation and
the IRS’s passive beamforming with a customized algorithm
initialization. A significant performance improvement on the
link rate can be observed compared to the cases without
IRS. The authors also show that there exists an optimal size
for grouping to maximize the achievable rate. The authors
in [132] focus on the phase error in channel estimation, which
brings the ambiguity in the IRS’s phase tuning. The composite
channel assisted by an IRS with N reflecting elements can
be modeled as Nakagami fading channel. Theoretical analysis
reveals that the average received SNR still grows with N 2 and
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
the error probability performance is robust against the phase
errors. The authors in [133] quantify the negative effect of
pilot contamination in channel estimation due to intra/inter-
IRS interference. An asymptotical analysis on the spectrum
efficiency of multi-IRS-assisted uplink transmissions reveals
that the achievable spectrum efficiency is limited by the
effect of pilot contamination and intra/inter-IRS interference
even with infinite reflecting elements. The above analyses
with finite phase resolution [128], inexact CSI [129]–[131],
estimation and quantization errors [132], [133] verify that the
practical imperfections can be generally counteracted by using
a larger-size IRS. The previous analyses in [106], [107] reveal
another countermeasure, i.e., dividing larger IRS into multiple
smaller IRSs, to enhance reliability and fight against channel
estimation errors. All these studies provide useful guidelines
for the IRS’s practical implementations in wireless systems.
2) Multi-user or Multi-cell Coordinated Communications:
The previous part reviews and verifies the potential perfor-
mance enhancement of point-to-point wireless communica-
tions by deploying IRS in the propagation environment. The
SNR or capacity maximization problems can be naturally
extended to the MU scenarios. However, the solution methods
will become more involved due to the resource competition
and interference among different users.
The authors in [134] present an efficient design for sum-
rate maximization in IRS-assisted downlink MISO communi-
cations, subject to the AP’s power budget constraint. Each
user is given a minimum data rate requirement. The sum-
rate maximization problem is firstly simplified by using the
zero-forcing (ZF) transmission scheme and then following
an iterative procedure to optimize the transmit power and
the phase shift matrix. This is achieved by combining the
alternating maximization with the majorization-minimization
(MM) method. The sum-rate maximization in a similar model
is also studied in [135] and solved heuristically by the alter-
nating optimization framework. In the ideal case with infinite
phase resolution, the authors in [134] show that the system
throughput can be increased by at least 40%, without addi-
tional energy consumption. The optimal solution to the sum-
rate maximization relies on the exact CSI, which is difficult to
obtain for large-size IRS. Considering a similar IRS-assisted
MU MISO system, the authors in [136] extend the sum-
rate maximization problem with perfect CSI to the case with
imperfect CSI. For perfect CSI case, the active and passive
beamforming strategies can be obtained by using the fractional
programming technique, which can be extended to the case
with imperfect CSI based on a stochastic successive convex
approximation (SCA) method. Simulation results reveal that
the proposed algorithm performs quite well when the channel
uncertainty is smaller than 10%. To avoid frequent channel
estimation, the authors in [137] propose a two-timescale
algorithm to maximize the sum-rate for an IRS-assisted MU
MISO downlink system. The IRS’s phase shift matrix is firstly
optimized based on the slow-varying statistical CSI, while
the AP’s transmit beamforming is then optimized according
to the instantaneous CSI. Besides, different from the popular
SDR-based algorithms [116], the authors in [137] leverage
the penalty dual decomposition technique that allows parallel
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
16
updates of the optimization variables. Hence, the proposed
algorithm can significantly reduce the channel training over-
head, computational time and complexity for optimizing the
IRS’s passive beamforming. Our previous observation in [100],
[101] reveals that the capacity gain using IRS becomes more
significant for asymmetric user rates. In particular, the capacity
gain can be achieved by deploying IRS to assist NOMA
transmissions to cell-edge users simultaneously with the other
users closer to the BS [138]. Hence, we expect that the IRS-
assisted sum-rate maximization will greatly improve the users
with weak channel conditions. However, it usually leads to
a non-convex design objective due to the interference and
couplings among different users.
The design objectives of sum-rate maximization in different
network scenarios are intrinsically fighting against the resource
competition or interference among different users. The use
of the IRS can make the wireless propagation channels more
flexible to control and thus easier for interference suppression.
Considering a typical MU MIMO system, the authors in [139]
show the feasibility of constructing multiple interference-free
beams by using the IRS with a large number of passive
elements. Analysis shows that a single set of optimal beam-
forming weights can form multiple interference-free beams
for multi-stream MIMO transmissions. The authors in [140]
investigate users’ interference in a more complicated multi-
cell multi-cast downlink system. The information destined
to different groups of users are independent and different.
Hence, there exists inter-group and inter-cell interference.
To maximize the sum-rate of all groups, a concave lower
bound of the objective function is firstly derived and then the
alternating optimization is employed to update iteratively the
BS’s precoding matrix and the IRS’s passive beamforming.
To reduce the computational complexity, the MM method
in [134] is adopted to derive a closed-form solution in every
iteration. The simulation results demonstrate that the sum-rate
can be improved by more than 100% when assisted by an
IRS with only 8 scattering elements, comparing to a massive
MIMO system with 256 antennas at the BS. To mitigate inter-
cell interference, the authors in [141] deploy the IRS at the
cell boundary of a multi-cell system to assist the downlink
MIMO transmissions to cell-edge users. The maximization
of weighted sum-rate is solved with a similar alternating
optimization as that in [140]. The BS’s active precoding and
the IRS’s passive beamforming are iteratively optimized by
using the block coordinate descent (BCD) algorithm. Note that
it is difficult to guarantee individual users’ rate requirements
in a sum-rate maximization problem. Though we can add in-
dividuals’ rate constraints as that in [134], [135], an improper
or uneducated setting for the users’ minimum rate requirement
may drastically bring down the sum-rate performance, or even
make the design problem infeasible.
The difficulties in sum-rate maximization can be resolved
by considering user fairness as the critical performance metric
for performance maximization in MU networks, which in
general can be formulated as max-min problems. The authors
in [142] maximize the minimum signal-to-interference plus
noise ratio (SINR) of all users in an IRS-assisted NOMA
system by jointly optimizing the BS’s power allocation and
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
the IRS’s passive beamforming. Simulation results show that
the IRS with 1-bit phase resolution improves the max-min
rate by 20% compared to that of traditional NOMA. Similarly,
the authors in [143] maximize the minimum user-rate in the
wireless system assisted by a set of distributed IRS units.
Each IRS unit has a separate signal processing unit and is
connected to a central processing unit that coordinates the
behaviors of all the IRS units. A user assignment scheme
between each user and the IRS is proposed to improve the
minimum user-rate. The optimal user assignment scheme can
be effectively found by solving classical linear assignment
problems defined on a bipartite graph. Numerical results
show that the proposed user assignment scheme is close to
optimum both under LOS and scattering environments. The
maximization of the minimum weighted SINR in a multi-cell
MISO downlink system is studied in [144], where an IRS is
dedicatedly deployed to suppress inter-cell interference and as-
sist information transmission of cell-edge users. With the fixed
passive beamforming at the IRS, the BSs’ transmit powers are
optimized by second-order-cone programming (SOCP). The
update of the IRS’s passive beamforming can be achieved by
using SDR and SCA methods. Numerical results show that
the proposed algorithm can achieve significant performance
gain, i.e., the minimum SINR can be increased by over 150%
with the BS’s transmit power at 35 dBm, compared to the
conventional case without IRS. The common observation from
the above works in [134], [135], [137]–[144] is that they all
consider interference constrained systems, where the spectrum
access of different users introduces mutual interference to each
other and thus limits the sum-rate performance. This also
complicates the algorithm design due to the users’ couplings.
Different from the above interference-limited networks, the
authors in [148] integrate IRS to an orthogonal frequency divi-
sion multiple access (OFDMA) based MU downlink system.
A joint optimization of the IRS’s passive beamforming and
OFDMA resource block (RB) as well as power allocations
is proposed to maximize the minimum user rate. By using
a dynamic passive beamforming scheme, the IRS’s reflection
coefficients are allowed to dynamically change over different
time slots. In each time slot, only a subset of the users will be
selected and served simultaneously, thus achieving a higher
passive beamforming gain. Numerical results show that the
dynamic passive beamforming outperforms the fixed passive
beamforming scheme. The performance improvement becomes
larger as the size of scattering elements increases. To this point,
all the above works assume continuously controllable phase
variables at the IRS. The analysis of performance degradation
with practical low-resolution IRS is desirable for evaluating
the robustness and reliability in practice.
The authors in [145] use a low-resolution IRS and de-
sign a symbol-level precoding scheme for MU MISO/MIMO
downlink system to minimize the worst-case symbol-error-
rate (SER). The discrete phase shifts are firstly relaxed as
continuous design variables and optimized by the Riemannian
conjugate gradient algorithm. Then, the low-resolution pre-
coding vector is obtained by direct quantization. As a special
case, the branch-and-bound method is proposed to reduce the
quantization error for the 1-bit symbol-level precoding vector.
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
17
Similarly, the authors in [147] employ the low-resolution IRS
to enhance the sum-rate in MU MISO downlink system. A
good sum-rate performance gain can be achieved by using the
IRS with a reasonable size and a small number of discrete
phase shifts. Considering weighted sum-rate maximization in
a similar MU MISO system, the authors in [146] propose
three iterative methods to optimize the discrete phase levels
of the IRS’s scattering elements in addition to the BS’s
beamforming optimization, based on the fractional program-
ming method. Numerical results show that the IRS with 2-
bit phase resolution achieves sufficient capacity gain with
only a small performance degradation. Similar observations
and implications are also made in [128], [130]. Note that the
aforementioned works all assume perfect or static CSI of all
users. In fact, the channel estimation involving passive IRS
becomes more complicated than before due to a large size of
passive reflecting elements. Besides, the channel estimation
in a dynamic environment is inevitably contaminated by error
estimates and thus leading to channel uncertainty, which has
not been fully investigated for the IRS-assisted MU scenarios
in the current literature.
Summary: In this part, we have reviewed the potentials of
using the IRS to improve transmission performance in terms
of SNR or data rate for both point-to-point communications
and multi-group/multi-cell MU cases. A summary of existing
works on SNR or capacity maximization is listed in Table V.
In particular, we have discussed the applications of the IRS
under different communication models, e.g., MISO, MIMO,
OFDM, NOMA, mmWave, and multi-cell systems. Different
system models are illustrated and compared in Fig. 9. The
performance maximization of IRS-assisted wireless systems
in different scenarios is typically formulated into a joint
optimization problem of the IRS’s passive beamforming and
the BS’s transmit beamforming or power allocation strategy.
Along this main line of research, some special cases are
also discussed, including the phase shift optimization for the
non-ideal IRS with low phase resolution, or with incomplete
or uncertain channel information. We notice that the main
solution methods are based on a simple alternating opti-
mization framework, which can guarantee the convergence
to sub-optimal solutions. However, comparing the optimum,
the performance loss by using the alternating optimization is
not known exactly and seldom characterized in literature. By
developing more sophisticated algorithms in the future work,
we envision that the IRS-assisted wireless systems can achieve
a higher performance gain than that in the current literature.
B. Power Minimization or EE/SE Maximization
Besides SNR and rate maximization, the IRS-assisted wire-
less networks can also help minimize the BS’s transmit power
or maximize the overall EE/SE performance. The IRS can con-
figure wireless channels in favor of information transmission
between transceivers. This results in a more energy-efficient
communication paradigm, e.g., the BS can maintain the same
transmission performance with a reduced power consumption.
As such, IRS-assisted communications can be envisioned as a
green technology for future wireless networks.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

18

TABLE V: SNR or Capacity Maximization in IRS-assisted Wireless Networks

Reference Optimization System Model Resolution Design Variables Main Results

[116] Max. signal MISO downlink Continuous Transmit and passive The IRS with size N achieves a total beamforming
power beamforming gain of N 2
[118] Max. data rate MISO downlink Continuous Transmit and passive Achieve a higher data rate with reduced computa-
beamforming tional complexity
[119] Max. data rate MISO OFDM Continuous Transmit power alloca- Effective in boosting the data rate of a cell-edge user
downlink tion, passive beamforming
[120] Max. data rate MISO CRN Continuous Transmit and passive IRSs can improve the data rate of the secondary user
downlink beamforming under both perfect and imperfect CSI cases
[121] Max. signal MISO mmWave Continuous Transmit and passive The received signal power increases quadratically
power downlink beamforming with N
[122] Max. capacity MIMO downlink Continuous Transmit covariance ma- Substantially increased capacity compared to MIMO
trix, passive beamforming channels without using the IRS
[123] Max. data rate mmWave hybrid Continuous Hybrid precoders, passive Achieve better BER and spectrum efficiency than the
MIMO OFDM beamforming conventional approaches
[124] Max. capacity MIMO downlink Continuous Passive beamforming Capacity gain validated by simulations
[126] Max. sum rate Two-way full- Discrete Transmitters’ precoders, Capacity gain is close to that of a full-duplex relay
duplex MIMO passive beamforming with transmit power at −40 ∼ −35 dBm
[128] Max. capacity MISO downlink, Discrete Discrete phase shifts 2-bit phase quantizer is sufficient to ensure capacity
statistical CSI degradation of no more than 1 bit/s/Hz
[129] Max. data rate SISO uplink Discrete Discrete phase control Significant rate improvement is achieved by a low-
complexity successive refinement algorithm
[130] Max. data rate MISO downlink, Discrete Discrete phase shifts Moderate to high phase resolution approximates the
unknown CSI optimum with continuous phase and perfect CSI
[131] Max. data rate SISO OFDM Continuous Transmit power alloca- An optimal grouping size exists to maximize the
downlink tion, passive beamforming achievable rate
[134] Max. sum-rate MU MISO down- Continuous Transmit power alloca- Throughput increased by at least 40%, without re-
link tion, passive beamforming quiring additional energy consumption
[135] Max. sum rate MU MISO down- Continuous Transmit power alloca- Significant performance gain achieved over the ran-
link tion, passive beamforming dom phase shift and the conventional ZF methods
[137] Max. sum rate MU MISO with Discrete Transmit precoders, dis- Significantly reduce the channel training overhead,
statistical CSI crete phase shifts computational time and complexity for optimizing
the IRS’s passive beamforming
[138] Max. sum rate MU MISO NO- Discrete, Discrete phase shifts Performance gain of IRS-assisted NOMA becomes
MA downlink binary significant for asymmetric user rates
[139] Min. MU MIMO Continuous Passive beamforming A single set of weights can form multiple
interference downlink interference-free beams for MIMO transmission
[140] Max. sum-rate MU multi-cell Continuous Precoding matrix, passive Improved EE/SE performance over conventional
MISO downlink beamforming massive MIMO systems
[141] Max. weighted MU multi-cell Continuous Precoding matrix, passive Significant performance gain is achieved over the
sum-rate MIMO downlink beamforming conventional counterpart without the IRS
[142] Max. min SNR MU SISO NO- Continuous Power allocation, passive 1-bit phase resolution improves the min rate by 20%
MA downlink beamforming compared to that of traditional NOMA
[143] Max. min rate MU uplink Continuous User-IRS assignment, pas- The user assignment performs close to the optimum
sive beamforming in both LOS and scattering environments
[144] Max. min SINR Multi-cell MISO Continuous Transmit and passive Min SINR can be increased by over 150% with the
downlink beamforming BS’s transmit power at 35 dBm
[145] Min. worst- MU MIMO Discrete, Precoding matrix, discrete Enhanced SER performance can be achieved
case SER downlink binary phase control
[146] Max. weighted MU MISO down- Discrete Transmit beamforming, The IRS with 2-bit resolution achieves sufficient
sum-rate link discrete phase control capacity gain
[147] Max. sum-rate MU MISO down- Discrete Transmit and passive A good performance gain achieved by using the IRS
link beamforming with a few discrete phase shifts
[148] Max. min rate MU OFDMA Continuous RB and power allocation, The IRS-assisted system achieves massive MIMO
downlink passive beamforming gains with a fewer number of AP’s antennas

Focusing on an IRS-assisted MISO downlink scenario as
that in [116], the authors in [149] minimize the AP’s transmit
power under individual users’ SINR constraints by jointly
optimizing the AP’s transmit beamforming and the IRS’s
passive beamforming strategies. Following SDR procedure
and alternating optimization, the AP’s transmit beamforming
can be efficiently optimized by solving SOCP, and the opti-
mization of the IRS’s passive beamforming is degenerated to
a conventional relay beamforming optimization problem. By
an asymptotic analysis with an infinite number of scattering
elements, the AP’s transmit power can be scaled down in the
order of 1/N 2 without compromising the SNR at the receiver.
Numerical results verify that the AP’s transmit power can be
reduced by more than 55% for the users far away from the AP
(e.g., 50 meters). The authors in [150] focus on an IRS-assisted
MISO broadcasting system and derive a lower bound of the
BS’s minimum transmit power, which is much lower than the
cases without IRS. Moreover, the BS’s transmit power can
approach the lower bound as the number of scattering elements
increases. Transmit power minimization is also considered
in an IRS-assisted NOMA downlink system [151]. The joint
optimization the active and passive beamforming is a highly

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
intractable bi-quadratic program, which is firstly relaxed via
SDR and then solved by the difference-of-convex (DC) algo-
rithm. Simulation results demonstrate that the AP’s transmit
power can be reduced by more than 8 dB when using an IRS
with 50 scattering elements. It is obvious that [149]–[151]
focus on point-to-point scenarios with simplified assumptions,
e.g., fully controllable phase shifts, static and known CSI.
The proposed design problems are expected to consider more
general cases with multiple users in a dynamic environment.
The transmit power minimization for MU MISO downlink
system is investigated in [152], where the users are grouped
into different clusters and NOMA is employed in each cluster
to enhance information transmission. An effective SOCP-
based alternating direction method of multipliers (ADMM) is
proposed to optimize the BS’s transmit beamforming and the
IRS’s passive beamforming. A ZF-based sub-optimal algorith-
m is also proposed to reduce the computational complexity.
The simulation results demonstrate significant performance
gain over the conventional SDR-based algorithms, e.g., [116],
[124], [144], [151]. To compare multiple access scheme for
MU downlink systems, the authors in [153] evaluate the
minimum transmit powers required by different schemes.
Assuming perfect CSI, the transmit power in either scheme
is minimized by alternating optimization. The comparison
reveals that the IRS-assisted NOMA may perform worse than
that of the IRS-assisted time division multiple access (TDMA)
for users closer to the IRS. To achieve NOMA gain over
TDMA, it is preferred to pair users with asymmetric rates
for NOMA downlink transmissions. Note that the BS in MISO
downlink systems typically has constant power supply. Hence,
the minimization of BS’s transmit power may not be an urgent
need. In fact, the BS’s energy efficiency can be a more critical
design objective, which represents the transmission capability
of the network.
Instead of minimizing transmit power, the authors in [154]
maximize the energy efficiency by jointly optimizing the
IRS’s passive beamforming and the AP’s power allocation
over different users, subject to the AP’s maximum power
and the users’ minimum QoS constraints. The total power
consumption includes the transmit power, the constant circuit
power, as well as the IRS’s power consumption, which relates
to the size and implementation of the reflecting elements.
In particular, a finer phase resolution or a larger size of
scattering elements implies a higher power consumption. Let
PIRS denote the IRS’s power consumption. The authors
in [154] present a simple linear power consumption model,
i.e., PIRS = N P (b), where N denotes the number of
scattering elements and P (b) is the power consumption of each
phase shifter with b-bit resolution. Typical power consumption
values of P (b) are 1.5, 4.5, 6, and 7.8 mW for 3-, 4-, 5-,
and 6-bit resolution phase shifting, respectively. A gradient
descent method is firstly used for optimizing the IRS’s phase
control, and then the transmit power allocation is optimized
by a fractional programming method. The simulation results
in a realistic environment show that the IRS-assisted system
can provide up to 300% higher energy efficiency than that of
the conventional multi-antenna AF-relay communications. The
authors in [155] compare the performance between the IRS-
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
19
assisted system and the conventional decode-and-forward (DF)
relay communications. The achievable rates for both cases are
maximized by optimizing the transmit power and the size of
the IRS. The main observation is that a large-size IRS is
needed to achieve better performance than that of DF-relay
communications, in terms of EE/SE performance. Though a
power consumption model is proposed for the IRS, the IRS’s
power budget constraint is not considered in the optimization
framework. Instead, the IRS is implicitly assumed to be always
online and capable of phase tuning.
Prior works mostly assume infinite phase resolution for the
IRS, which however is practically difficult to realize due to
the hardware limitation. Considering a more practical case,
the authors in [156] and [158] minimize the AP’s transmit
power in a downlink MISO system, assisted by an IRS with
finite phase resolution. The feasible discrete phase shifts can
be obtained by quantization projection from the optimized
continuous phase values, similar to [145]. Analytical results
show that a practical IRS with discrete phase shifts can still
achieve the same power scaling law as that with continuous
phase shifts, the finding similar to [116]. More interestingly,
the performance loss due to quantization errors is shown to be
irrelevant to N , the number of reflecting elements, while only
dependent on the IRS’s phase quantization level b. Numerical
results reveal that the discrete phase shifts with b = 2 or b = 3
are sufficient to achieve the close-to-optimal performance,
which is consistent with the former observation in [128], [130],
[146]. The optimization solutions to the single-user MISO
system in [156] can be extended to an MU case in [158].
However, the quantization projection used in [156] may lead
to higher performance degradation in the MU case due to
the severe co-channel interference. A special case with the
1-bit phase resolution is studied in [157]. The optimization of
phase shift matrix and power allocation for each user follows
a similar alternating optimization method. The IRS’s perfect
phase control for both infinite and finite phase resolutions
depends on the knowledge of exact CSI, which however is
practically challenging to obtain due to the lack of signal
processing capability at the IRS and the large size of passive
scattering elements.
Summary: In this part, we have reviewed the use of the
IRS in wireless networks to minimize the transmit power
or maximize the energy efficiency. The literature reveals an
important power scaling law showing that the BS’s power
consumption can be scaled down in the order of 1/N 2 without
compromising the SNR at the receiver. The similar power
scaling law still holds for a practical implementation of the
IRS with low phase resolution. The power saving of IRS-
assisted systems becomes more significant for wireless users
far away from the transmitter. Though significant performance
improvement can be verified by numerical results and simula-
tions, we observe that the research focuses of almost all papers
in literature are limited to the joint optimization of active and
passive beamforming under different network scenarios. In
fact, the overall performance gain can be better explored in
the future work if the size of the IRS’s scattering elements,
the orientation of the IRS tiles, their partitions and grouping
strategies, etc., are all taken into account, in combination with
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

20

TABLE VI: Power Minimization or EE Maximization in IRS-assisted Wireless Networks

Reference Optimization System Model Resolution Design Variables Main Results

[149] Min. transmit MISO downlink Continuous Transmit and passive Transmit power can be scaled down in the order of
power beamforming 1/N 2 without compromising the SNR at the receiver
[150] Min. transmit MISO downlink Continuous Transmit and passive A lower bound is derived for the BS’s minimum
power broadcasting beamforming transmit power
[151] Min. transmit MISO NOMA Continuous Transmit and passive Verified effectiveness and superiority of using IRS to
power downlink beamforming reduce the total transmit power
[152] Min. transmit MISO NOMA Continuous Transmit and passive The IRS-assisted ZF scheme outperforms the SDR-
power downlink beamforming based algorithm when N is large
[153] Min. transmit MU SISO NO- Discrete Transmit power alloca- NOMA gain can be achieved by pairing users with
power MA downlink tion, discrete phase shifts asymmetric rates
[154] Max. energy effi- MISO downlink Continuous Power allocation, passive IRS-assisted system provides up to 300% increase in
ciency beamforming energy efficiency than AF relay communications
[156] Min. transmit MISO downlink Discrete Transmit beamforming, A discrete IRS can achieve the same power gain as
power discrete phase control that with infinite phase resolution
[157] Max. energy effi- MU MISO down- Discrete, Transmit beamforming, 1-bit resolution IRS significantly improves energy ef-
ciency link binary discrete phase control ficiency compared to relay-assisted communications

the transceivers’ access control, user association, information
encoding, transmit scheduling, QoS provisioning, etc.
More specifically, we notice that there are a few potential
research problems that are not fully explored in the current lit-
erature and thus can be left for future study. Firstly, the power
budget constraint of a practical IRS is barely mentioned in the
literature. The EE/SE performance maximization relies on the
characterization of total power consumption in IRS-assisted
wireless networks. The power consumption of a practical IRS
is not negligible and can be modeled by a linear function of its
size and tuning resolution [154]. However, most of the current
researches consider self-sustainable IRS with sufficient energy
supply, which may lead to over-optimistic conclusions. A
preliminary work in [159] introduces the IRS’s power budget
constraint in a simple point-to-point MISO downlink system,
which can be fulfilled by RF energy harvesting from the
AP’s signal beamforming. It is worthy further investigation
in future research due to the importance of energy budget
constraint in energy constrained IoT networks. Secondly, we
may expect more research effort dedicated to performance
maximization of MU systems under different multiple access
schemes. The majority of current research works focus on
interference-limited MU systems, where IRS is employed to
assist spectrum sharing among different users and improve
spectrum efficiency. The research of IRS-assisted multi-user
access and performance maximization in different MAC proto-
cols is very limited and should be explored more in the future.
For example, we only observe one work in [148] focusing on
IRS-assisted RB allocation in an OFDMA system. Besides, we
notice that IRS-assisted systems face the common challenges
of channel estimation and the design approaches under incom-
plete or uncertain channel information. The authors in [137]
show their effort along this research direction by designing a
two-timescale sum-rate maximization algorithm that reduces
the need for channel estimation. While a large majority of the
existing works assume perfect channel information and focus
on static maximization problems.
C. IRS-assisted Physical Layer Security
The IRS’s wave manipulation has the flexibility of simulta-
neously creating enhanced beams to an intended receiver and
suppressed beams to unintended receivers. This can be used to
enhance physical layer security in wireless communications.
The authors in [160] use IRS to defend against eavesdroppers.
In particular, a single-antenna eavesdropper is located in
the communication range between a multi-antenna legitimate
transmitter (LT) and a single-antenna legitimate receiver (L-
R). To prevent the eavesdropper from eavesdropping by, the
IRS placed near the LR can control the reflected signals to
maximize the achievable secrecy rate at the LR, which is
defined as the amount of information per time unit that can
be securely sent over a communication channel [161]. The
joint optimization of the LT’s transmit beamforming and the
IRS’s phase shift matrix is approximately solved by BCD and
MM algorithms, similar to that in [134], [140]. Simulation
results verify a significant improvement on the secrecy rate
comparing to the cases without IRS. Besides, it is shown that
it can be more efficient to enhance secrecy rate and energy
efficiency by deploying large-scale IRSs instead of increasing
the size of active antenna array at the transmitter. This idea
has been extended in the recent research works by considering
multiple antennas at the LR/eavesdropper [162], [163], rank-
one/full-rank communication channels [164], and secure THz
communications [165]. Besides secrecy rate maximization, the
authors in [166] minimize the transmit power subject to the
secrecy rate constraint at the LR, by optimizing transmit power
allocation and the IRS’s phase shift matrix in the alternating
optimization framework. The above research works assume
implicitly that the eavesdropper is distant to the LR, or the
eavesdropper has a worse channel condition, which is required
to ensure the effectiveness of secure communications.
Considering a similar IRS-assisted secure wireless system
to that in [160], the authors in [167] focus on a more
challenging scenario in which the eavesdropping channel is
better than the legitimate channel and they are also highly
correlated in space. This makes the achievable secrecy rate of
the system very limited. To maximize the achievable secrecy
rate, the optimal solution requires that the IRS’s reflections

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

21

Intelligent Reflecting Surface (IRS)

Base station (BS)

Information
exchange IRS controller

Passive
beamforming
Active
beamforming

End user d1 d1 dK
(a) IRS-assisted MISO downlink communications (b) IRS-assisted MU downlink communications

Distributed IRS

d1 dK d1 dK
(c) IRS-assisted MU MIMO downlink communications (d) MU downlink communications assisted by
distributed IRSs

BS1

BS2
Cell-edge
user
Group g1 gK
(e) IRS-assisted multi-group multi-cast downlink (f) IRS-assisted downlink communications to
communications cell-edge users

Fig. 9: Different system models for IRS-assisted wireless networks. (a) Basic model of IRS-assisted downlink transmission from
multi-antenna BS to one receiver, which can be a single-antenna end user, e.g., [116], [118], [119], [121], [128], [130], [149],
[151], [154], [156], or equipped with multiple receiving antennas, e.g., [122]–[124], [131]. (b) The IRS-assisted MU MISO
downlink transmissions, e.g., [134], [135], [137], [138], [146]–[148], [150], [157]. The special case with single-antenna BS
is studied in [131], [142], [153]. (c) The IRS-assisted downlink transmissions to multi-antenna receivers (MIMO), e.g., [139],
[145]. (d) Multiple-IRS-assisted MISO/MIMO downlink transmissions, e.g., [100], [106], [107], [120], [121], [143]. (e) IRS-
assisted multi-group multi-cast downlink transmissions, e.g., [140], [152]. (f) IRS-assisted downlink transmissions to cell-edge
users, e.g., [141], [144].

and the LT’s beamforming signals are destructively added
at the eavesdropper. This idea is then extended to a more
general scenario in which multiple eavesdroppers and LRs
coexist in the system [168], where the authors maximize the
users’ minimum-secrecy-rate considering both continuous and
discrete phase shifts at the IRS. This problem is approximately
solved by the alternating optimization and path-following
algorithms in an iterative manner. In practice, it is generally
difficult to know the number, location, and channel conditions
of eavesdroppers, and thus more practical solutions based on
incomplete environmental information can be studied in future
work. Another drawback of the above works is that the LT
has very limited space to fight against the eavesdropper. It is
obvious that the achievable secrecy rate is constrained by the
network topology and channel conditions.
Besides the LT’s transmit beamforming, the authors in [169]
allow the LT to actively inject noise-like jamming signals to
the channel, and further optimize the LT’s jamming signals to
enhance the achievable secrecy rate. By incorporating jamming
signals into transmit beamforming, the secrecy rate is shown
to be significantly improved compared to the conventional
methods without using the IRS and/or jamming beamforming.
The authors in [170] consider a similar model and optimize the
covariance matrix of artificial noise to deliberately impair the
eavesdropper’s channel, along with optimization of the BS’s
signal beamforming and the IRS’s phase shift matrix based
on SCA and SDR techniques. The numerical results show
that the average secrecy rate can be improved by over 20%,
compared to the case without using dedicated artificial noise.
The authors in [20] consider different approach to maximize
the secrecy rate, by optimizing the number of IRS tiles (namely
hypersurface tile) and their orientations to create desirable

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
Intelligent Reflecting Surface (IRS)
Transmit
beamforming
IRS controller
Passive
beamforming
Legitimate transmitter (LT)
Eavesdropper
Legitimate receiver (LR)
Jamming
beamforming
Friendly jammer
Fig. 10: An IRS-assisted secure communication system with
a friendly jammer.
phase changes. In this way, the received signal power at the
eavesdropper becomes very small even if the eavesdropper is
located between the LT and LR. Though this is an effective
solution to prevent eavesdropping, it requires the implemen-
tation of the hypersurface tile and thus its applications are
limited to some particular scenarios such as smart offices and
houses. All aforementioned works generally rely on the LT’s
optimal control (e.g., artificial noise injection and transmit
beamforming) combined with the IRS’s passive beamforming
to improve the secrecy performance. Different from them, the
authors in [171] introduce an interesting idea by using a third
device, namely a friendly jammer, to cooperate with the LT and
to fight against multiple eavesdroppers, as shown in Fig. 10.
The secrecy rate maximization in this case requires a joint
optimization of the LT’s transmit beamforming, the friend’s
jamming beamforming, and the IRS’s passive beamforming.
The solution method follows a similar approach as those
in [160], [168], by leveraging the alternating optimization and
SDR approaches.
The secrecy rate maximization merely aims at preventing le-
gitimate transmissions from being deciphered or eavesdropped
by an illegitimate user. However, this may not be enough for
some cases where exposures of transmission activities, loca-
tion and movement of transmitters are sensitive information
to the end users. This calls for covert communications that
can provide stronger protection by hiding the presence of
legitimate transmissions [172]. The authors in [173] propose
the use of the IRS to enhance communication covertness, by
leveraging IRS to reshape undesirable propagation conditions
and thus avoid information leakage. A joint optimization of
the LT’s transmit power and the IRS’s passive beamforming
is proposed to maximize the achievable rate satisfying the
covertness requirement. The numerical results demonstrate a
significant increase in covert rate compared to the conventional
cases without using the IRS. However, the efficacy of this work
is based on the knowledge of perfect channel conditions.
Summary: In this section, we have reviewed emerging
applications of IRS for physical layer security. In particular,
IRS can be used as a very effective tool to prevent wireless
eavesdropping attacks by simultaneously controlling the trans-
missions at the LT and the reflection at the IRS. As a result, the
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
22
achievable secrecy rates obtained by the IRS-assisted systems
can be significantly improved compared with the convention-
al methods only relying on the LT’s transmission control.
Many simulation results verify the improvement of secrecy
performance of IRS-assisted systems. This paves the way for
a flourish of new research problems relating to physical layer
security issues in the future wireless networks. However, there
are still some challenges which need to be addressed. For
example, how to simultaneously control the LT’s transmissions
and the IRS’s reflections in real-time systems, and how to
obtain the eavesdropper’s channel information for accurate
beamforming optimization are still major challenges for anti-
eavesdropper systems.
V. EMERGING USE CASES OF IRS IN WIRELESS
NETWORKS
As demonstrated in previous sections, the use of the IRS is
capable of bringing unprecedented performance enhancement
for future wireless systems by reconfiguring the previously un-
controllable wireless channels in favor of network performance
optimization. In particular, we have reviewed in previous
sections the performance analysis and optimization of IRS-
assisted wireless networks with different design objectives,
i.e., SNR/capacity maximization, transmit power minimiza-
tion, EE/SE performance maximization, and physical layer
security issues. However, the potentials of using the IRS in
wireless systems are not limited to the above aspects. It is
still developing and flourishing in various aspects far beyond
the aforementioned topics.
A. Deep Learning for IRS-assisted Systems
In Section IV, the joint active and passive beamforming
design problems are mostly formulated as a non-convex prob-
lem and typically solved with sub-optimum in the alternating
optimization framework. As the size of scattering elements
becomes large, e.g., up to hundreds [12], the optimization
problem usually has high computational complexity and thus
becomes difficult for practical implementation in a dynamic
radio environment. Different from optimization methods, the
authors in [174] propose a deep learning (DL) method for
efficient online reconfiguration of the IRS in a complex indoor
environment. It is based on the fact that the IRS’s optimal
phase configuration depends on the receiver’s location to
maximize the received signal strength. To save time for online
optimization, the deep neural network (DNN) is employed
to construct a direct mapping between the receiver’s location
and the optimal phase configuration. The offline training of
the DNN is based on a fingerprinting database recording
the optimal phase configuration at each receiver’s position.
Similarly, the authors in [175] propose a DL approach for the
IRS’s passive beamforming optimization, which is based on a
well-trained DNN to make real-time predictions. Simulation
results show that the DL approach achieves close-to-optimal
performance with significantly reduced time consumption
compared to the conventional SDR-based methods. However,
in the above learning based approaches, the optimal phase
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
configuration in offline training is still achieved by exhaustive
search or the alternating optimization method.
The authors in [176] propose a novel deep reinforce-
ment learning (DRL) approach to address the secrecy rate
maximization problem. DRL provides a mechanism to build
knowledge from scratch and make autonomous decisions to
improve network performance by continuously interacting
with the environment [177]. It can be very effective for
complicated systems with diverse user requirements and time-
varying channel conditions. However, a training process is still
required to attain the decision intelligence. The DL approaches
is also applied to the IRS’s channel estimation problem, which
typically incurs huge training overhead due to the large number
of reflecting elements. The authors in [178] focus on channel
estimation of an IRS-assisted massive MIMO system and
propose a DL solution to learn how to optimally interact with
the incident signal. The authors in [179] construct an inter-
pretable neural network for learning the optimal configuration
of reconfigurable reflecting elements, which are modeled as
the neural nodes in different layers of a back-propagating
neural network. By a training period, the neural network can
learn the propagation characteristics and adapt to facilitate the
signal transmissions. Numerical evaluation shows that it can
minimize the number of IRS tiles required for serving one
transceiver pair. Practically, the training of DL approaches
can be time consuming and unreliable, which may prevent
its implementations in real systems.
B. IRS-assisted Wireless Power Transfer
The IRS’s passive beamforming can be designed to enhance
the received signal strength at an information receiver (IR).
This approach can also improve the efficiency of wireless
power transfer to an energy receiver (ER). The authors in [180]
consider a point-to-point MISO system for wireless energy
transfer. The power beacon station’s active beamforming and
the IRS’s passive beamforming are jointly optimized to max-
imize the signal power at the receiver. This incurs a trade-
off between the size of active antenna array at the beacon
station and the size of passive reflecting elements at the IRS.
Considering multiple ERs, the authors in [181] maximize the
weighted sum power at the ERs by jointly optimizing the BS’s
transmit beamforming and the IRS’s phase shifts. A similar
model is studied in [182]. The design objective is to maximize
the ERs’ minimum power subject to the IRs’ SINR constraints
and the AP’s power constraint. A high-quality solution is
obtained by SDR and alternating optimization.
The authors in [183] consider an IRS-assisted simultaneous
wireless information and power transfer (SWIPT) system, in
which a multi-antenna BS communicates with several multi-
antenna IRs, while guaranteeing the ERs’ energy harvesting
requirements. The authors firstly formulate a weighted sum-
rate maximization problem by jointly optimizing the BS’s
transmit precoding and the IRS’s phase shifts. The classic
BCD algorithm is adopted to find a near-optimal solution,
which is much better than those of baselines, i.e., with fixed
phase or without using the IRS. A special case of the IRS-
assisted SWIPT is studied in [184], where two users firstly har-
vest energy and then transmit information to an AP following
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
23
the TDMA protocol. To maximize the minimum throughput,
the authors in [184] propose a joint optimization of the users’
time and power allocations, as well as the IRS’s passive
beamforming in wireless energy and information transfer. The
numerical results reveal that on average the IRS-assisted case
can achieve over 74% throughput increase compared to the
case without IRS. The authors in [185] consider a more general
MU MISO scenario, where the IRS firstly assists downlink
energy transfer to the users and then enhances the users’ uplink
information transmissions in the TDMA protocol. A sum-rate
maximization problem is formulated by jointly optimizing the
user’s transmission scheduling, the IRS’s phase shift matrices
for energy and information transfer. Simulation results verify
that the proposed scheme can improve the sum-rate by 350%
compared to the case without IRS. The authors in [186] use a
set of distributed IRSs to assist SWIPT from a multi-antenna
AP to multiple IRs and ERs. Instead of sum-rate maximization,
the authors in [186] focus on the AP’s transmit power mini-
mization by jointly optimizing the AP’s transmit beamforming
and the IRSs’ passive beamforming, subject to the IRs’ SINR
constraints and the ERs’ energy constraints. Simulation results
demonstrate the significant performance gains achieved over
benchmark schemes, e.g., the AP’s transmit power can be
reduced by more than 50% with only 30 scattering elements
in the demonstrated setup.
C. IRS-assisted UAV Communications
The capacity maximization problems in [122] and [131]
can be extended to the emerging UAV networks. In [187],
the authors introduce an IRS-assisted UAV communication
network in which the on-building IRS is used to enhance
communication quality from UAV to the ground user. A joint
optimization of the UAV trajectory and IRS’s passive beam-
forming is formulated to maximize the average achievable
rate. Given the UAV trajectory, the IRS’s phase shift is firstly
derived in a closed form. Then, with the fixed phase shifts,
the local-optimal trajectory solution can be derived by using
the SCA method. The authors in [188] employ the wall-
mounted IRS to enhance the channel between cellular BS
and UAVs, which previously suffers from poor signal strength
as the BS’s signal beamforming is generally optimized to
serve ground users. By controlling the IRS’s reflecting phase,
the signal gain at the UAV is characterized based on the
3GPP ground-to-air channel models as a function of various
deploying parameters, including the UAV’s height, the IRS’s
size, altitude, and distance to the BS. Thus, the maximum
signal gain can be achieved by optimizing the IRS’s location,
altitude, and distance to BS. Numerical results show that a
significant signal gain can be achieved for UAVs even with a
small-size IRS, e.g., the signal gain quickly jumps to 20 dB
when UAVs fly over 30 meters above the BS.
The authors in [85] deploy aerial IRS to enjoy full-angle
reflection and LOS channel conditions. The aerial IRS can
be carried by balloons or UAVs and used to enhance signal
coverage of a cellular network. The maximization of the worst-
case SNR in a coverage area on the ground is formulated
as an optimization of the transmit beamforming, the place-
ment, and phase shift matrix of the aerial IRS. A similar
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
scenario is studied in [189] where the UAV-carried IRS is
used to enhance the transmission performance of mmWave
communication networks. The UAV-carried IRS is also capable
of energy harvesting from the mmWave signals to sustain
its operations. A reinforcement learning (RL) approach is
proposed to find the optimal policy, i.e., the best location of
UAV, that maximizes the average throughput of the downlink
transmissions. The simulation results show that the RL-based
approach can improve the network performance by 65%
compared to conventional schemes without learning capability.
D. IRS-assisted Mobile Edge Computing
MEC allows data and computation offloading to resource-
rich MEC servers. As such, the energy consumption and
processing delay can be potentially reduced at the end users
with insufficient computing capability. However, the benefit
of MEC is not fully exploited, especially when the link for
data and computation offloading is hampered. As the use of
IRS can enhance both the EE/SE performance, it can be a
promising technology to improve the MEC performance.
The authors in [190] propose an IRS-assisted MEC sys-
tem where multiple users report individuals’ data to the
AP for data aggregation. The offloading process is assisted
by the IRS’s passive beamforming to enhance the channel
conditions. The design target is to minimize the worst-case
mean-squared-error in data aggregation, which is solved by
a novel alternating difference-of-convex (DC) programming
algorithm. The authors in [191] use the IRS in a green edge
inference system, where the computation tasks at resource-
limited user devices can be offloaded to multiple resource-
rich BSs. To minimize the network power consumption in
both computation and uplink/downlink data transmissions, the
authors in [191] propose a joint optimization of the task
allocation among different BSs, each BS’s transmit and receive
beamforming vectors, the users’ transmit power, and the IRS’s
passive beamforming strategies. The authors firstly propose
a reformulation by exploiting the group sparsity structure of
the beamforming vectors, and then decouple the optimization
variables by a block-structured optimization approach. Instead
of the widely used SDR approach, a novel DC-function based
three-stage framework is introduced to solve the original prob-
lem with enhanced network performance. Numerical results
reveal that the proposed approach can reduce the overall power
consumption by around 20% compared to the conventional
SDR-based approach.
Summary: The IRS is a cutting-edge technology possessing
outstanding features expected to open new promising research
directions, which have never been seen before in wireless
communication networks. In this section, we have reviewed
some emerging applications of the IRS in wireless networks
including wireless power transfer, UAV communications, and
MEC. It can be clearly seen that by using the IRS, energy,
communications and computing efficiency of conventional
wireless networks can be significantly enhanced. However, all
results obtained so far are based on simplified models and ver-
ified through simulations. More proof-of-concept prototypes
are required to validate the IRS’s practical efficiency. Besides,
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
24
there are more potential applications that can be enhanced by
using the IRS, which can be studied further in the future work,
such as vehicular and maritime communications, satellite and
next-generation mobile networks.
VI. CHALLENGES AND FUTURE RESEARCH
DIRECTIONS
Different approaches reviewed in this survey evidently show
that IRS-assisted wireless networks can effectively enhance the
received signal power, extend the network coverage, increase
the link capacity, minimize the transmit power, suppress the
interference, and enable better security and QoS provisioning
to multiple users, etc., compared to non-IRS-assisted counter-
parts. However, there still exists some challenges, open issues,
and new research directions which are discussed as follows.
A. Challenges and Open Issues
1) Energy-efficient Channel Sensing and Estimation: The
IRS is composed of a large array of passive scattering ele-
ments which are typically interconnected and controlled by a
centralized controller, e.g., [40], [41], [51]. The superiority of
using the IRS relies on its reconfiguration of each scattering
element’s phase shift, according to the channel conditions from
the transmitter to its receiver. This requires the capability of
channel sensing and signal processing, which becomes very
challenging without dedicated signal processing capability at
the passive scattering elements. The channel estimation of an
IRS-assisted system is typically performed at one end point
of the communication process, e.g., the BS with higher com-
putation capability. Existing approaches for the IRS’s channel
estimation generally assume that only one scattering element
is active each time, while all the other elements are inactive,
e.g., [30], [180]. Such an element-by-element ON/OFF-based
channel estimation scheme is practically costly for a large-
scale IRS with massive scattering elements. In particular, the
IRS is not fully utilized as only a small portion of the scatter-
ing elements is active in each time. This degrades the channel
estimation accuracy and produces a long estimation delay. The
authors in [192] propose the optimal activation pattern for the
IRS’s channel estimation, which achieves one order lower esti-
mation variance compared to the ON/OFF-based methods. The
authors in [193] and [194] assume that all reflecting elements
are switched on. By setting specially designed phase shift
at each reflecting element, the IRS-enhanced uplink channel
can be estimated efficiently with the reduced length of pilot
sequence. More sophisticated signal processing algorithms
are also designed for channel estimation, e.g., [195]–[197],
aiming at achieving better accuracy or lower training overhead,
however they generally demand higher power consumption for
information exchange, signal processing, and computation. A
practically efficient and sustainable channel estimation is still
one of the key enabling technologies for IRS-assisted wireless
systems.
2) Practical Protocols for Information Exchange: General-
ly, the IRS’s channel sensing and estimation can be achieved
by overhearing a training sequence sent by the active transceiv-
er. Thus, information exchange between IRS and the active
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
transceiver is required to synchronize the overhearing and
pilot training. Information exchange also happens when a
transmission scheduling protocol is employed to coordinate
MU’s data transmissions. In this case, the IRS also needs to
synchronize with different transmission frames and reconfigure
its passive beamforming schemes according to different users’
channel conditions. A practical protocol is thus required for
the IRS to talk with the conventional transceivers. Information
exchange can be made easy for conventional transceivers
using a dedicated control channel. However, without sufficient
energy supply, it becomes more challenging for the passive
IRS to detect and decode the information from other active
transceivers. Hence, the design of an information exchange
protocol firstly has to be of extremely low power consumption
such that it is sustainable by wireless energy harvesting.
Secondly, it has to be cost-effective by minimizing the conflict
with or the alteration of the existing systems. A potential
design idea can be incurred by the IRS’s inherent sensing
capability [10]. In particular, it can be more energy-efficient
for the IRS to sense physical layer information, instead of
decoding MAC layer information bit streams. Therefore, the
information exchange can be made possible by modulating
the packet length or transmit power, so that the passive IRS
can sense the variations of signal power with low-power
consumption.
3) Reflection as a Resource for IRS-assisted HetNets: In
the future smart radio environment, the wireless networks
can be assisted by a distributed IRS system with individually
controlled IRS units due to the pervasive deployment of
reconfigurable metasurfaces on different objects. This implies
a challenging situation for the real-time allocation and op-
timization of different IRSs to serve multiple data streams
in dynamic and heterogeneous networks (HetNets). Conven-
tionally, individual transceivers can independently adapt their
operational parameters to the channel condition, which follows
some stochastic model and can be predicted or estimated via
a training process. However, with the IRS’s reconfigurability,
the radio environment itself becomes controllable and non-
stationary. Hence, it becomes more difficult for individual
transceivers to understand the CSI via training. This implies
a centralized coordination for the IRS-assisted networks, at
least for the distributed IRS units. This makes the wireless
environment tractable and controllable. Specifically, a joint
control mechanism is required for efficient allocation and
association of IRS units to serve multiple users simultaneously.
4) Agile and Light-weight Phase Reconfiguration: The
phase control of an individual scattering element has to be
coordinated with each other for effective beam steering. The
large size of the IRS’s scattering elements can make the overall
phase tuning more flexible, even with limited phase shifts at an
individual scattering element, e.g., [128], [130], [146], [156].
Such a flexibility comes with a cost. In particular, an efficient
algorithm is required to jointly control the phase shifts of
all scattering elements in a timely manner, according to the
dynamics of the radio environment. The increase in the size
of scattering elements also imposes great pressure to channel
estimation, making it even harder for efficient phase control.
Besides, the IRS’s phase control is strongly coupled with
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
25
the transmit control of the active transceivers. This makes it
more challenging for the design of an agile and light-weight
phase control algorithm with minimum energy consumption
and communication overhead. In the current literature, most
of the phase control algorithms are based on alternating
optimization method, that decouples the IRS’s phase control
and the conventional transmit control (e.g., power allocation,
transmit beamforming, and precoding matrix design) in sep-
arated sub-problems. Though this simplification can provide
a convergent solution to a sub-optimum, it inevitably incurs
large communication overhead and processing delay. It is also
very challenging to characterize the performance loss of the
convergent solution compared with the optimum.
B. Future Research Directions
Based on extensive literature review and the analysis of
common shortcomings of the current literature, we highlight
a few potential research directions for future exploration.
1) Learning Approach for Passive Beamforming: Different
from the alternating optimization commonly used in the liter-
ature, machine learning approaches can be more appealing for
the IRS to realize agile and light-weight phase control based
on locally observed information of the radio environment [26],
[175], [198]. This can help minimize the overhead of infor-
mation exchange between the IRS and active transceivers.
The large number of scattering elements and their sensing
capabilities further imply that rich information can be collected
during channel sensing, providing the possibility for data-
driven DL approaches [10], [174], [175], [199]. Furthermore,
the potential analog computation can be also envisioned to
realize ANNs via multi-layer metasurfaces [70], [179], which
potentially make the learning approach agile in computation
and light-weight without the need for information exchange.
However, current DL approaches are still facing many prac-
tical challenges, including the training overhead, stability and
adaptability issues. The design of DL approaches has to meet
the hardware constraints of IRS-assisted wireless systems,
such as limited computation and communication capabilities
of the passive scattering elements. For example, leveraging
RL approaches, a decision-making agent can be employed
at the IRS controller to adapt its phase configuration, solely
based on the observed system state (e.g., the perceived CSI
via its sensing capability) and the receivers’ feedback of its
phase configuration, e.g., [176], [189]. The system state can
be estimated by the IRS via sensing or overhearing the ACK
packets from the receiver to the transmitter. With specially
designed ACK packets, e.g., the ACK packets with different
time durations or transmit power, the channel sensing of IRS
can be made easier without energy consumption on decoding
the ACK packets. A similar idea has been used for information
exchange in wireless backscatter communications [200].
2) IRS-assisted D2D Communications: D2D communica-
tions technology is envisioned to connect billions of low-power
user devices. Different from the typically downlink transmis-
sions from multi-antenna AP to receivers, D2D communica-
tions become more decentralized and diverse, which brings
new research problems for IRS-assisted D2D communications.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
In one aspect, The IRS can be dynamically reconfigured
to enhance individual data link of D2D communications.
This requires highly efficient channel sensing and estimation
protocols, as well as agile phase reconfiguration algorithms.
The insufficiency of energy supply for the IoT devices implies
another difficult situation that demands minimized interactions
between the IRS and the IoT devices. In another aspect, the
distributed IRS units can be used to understand the system
profile by learning from a large amount of IRS-assisted trans-
missions in a spatial-temporal region [201]. The system profile
may include the information about the potential bottleneck
devices, the time-varying traffic pattern, the energy distribution
over the entire network, and the information for predicting and
diagnosing network failures. Such information can be further
used by the D2D networks to optimize the deployment and
settings of the IRS units, the IoT devices’ transmission control,
the placement of relay nodes, and the power beacon stations.
3) IRS-assisted mmWave and THz Communications: One
of the promising applications of the IRS is in the extended
coverage of 5G and beyond 5G communications. We expect
that mmWave 5G communication and future THz beyond 5G
communication will face with a critical issue of dead-spots
which will not be covered well because of the severe blocking
loss of such short-length waveforms. In such situations, the
IRS’s two salient EM properties of reflection and refraction
can be exploited to resolve the critical issue of dead-spots
when IRSs are deployed in between the base stations and end
users. For example, a user is located in the same side of the
serving BS, in which case the incident EM wave on the IRS
can be reflected toward the user, whereas if a user is in the
opposite side, then the incident EM wave can be refracted
through the IRS to reach the user with enhanced signal quality.
It is envisioned that the 3D deployment of IRSs with 5G and
beyond 5G wireless systems for eliminating such dead-spots
will be cost-effective, and the current massive MIMO and
mmWave technologies evolving will be integrated with the
IRS technology for the extended coverage eventually.
4) Using IRS in Smart Wireless Sensing: The current re-
search typically uses the IRS as an auxiliary way for enhancing
transmission performance of the existing transceivers. In fact,
each scattering element of the IRS can be individually phase-
tuned and thus showing different sensitivities to the incident
signals from different directions. This implies that the IRS can
be employed as an array of sensor devices that are configured
to passively monitor the radio environment [10]. Given wired
or wireless connections to a centralized IRS controller, all the
sensing information from different scattering elements can be
collected and analyzed jointly in an energy-efficient way. From
this view point, the use of the IRS as an array of smart sensors
will have rich applications in wireless sensing, e.g., indoor
positioning [96], [108] and human pose understanding [202].
As such, IRS-assisted wireless systems will not only enhance
communications but also bring the possibility of human-
network interactions, i.e., the communication performance and
user satisfaction can be even better by using IRS to understand
the human behavior or intention in wireless networks [203].
5) Trade-off between Array Gain and LOS Path Loss:
Using the IRS as passive relay provides the array gain N 2
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
26
thanks to no noise addition when N reflecting elements
are employed. In addition, if the reflecting element is of
sufficiently large size (e.g., 10λ × 10λ for the wavelength λ),
the IRS can act as the specular reflector like lens in which
case the path loss follows the “sum-distance" path loss, unlike
the active relay whose antenna size is on the order of λ,
resulting in the severe “product-distance" path loss. Therefore,
there exists a crucial trade-off between achieving a larger array
gain and guaranteeing the minimal LOS path loss because the
number of reflecting elements varies per unit area depending
on the size of the reflecting elements. Namely, to assure the
LOS path loss, we have to sacrifice the array gain whereas
the largest array gain can be achieved while compromising the
LOS path loss. The latter is due to the minimum physical size
of a reflecting element (like lens) that focuses the energy onto a
focal point depending on the distance between the IRS and the
receiver (i.e., focal point). Future research for characterizing
the trade-off, considering the 3D deployment of IRSs, will be
of paramount importance, in that the smart radio environment
can be fully utilized in terms of the density of IRSs and their
sum gain, normalized by the implementation cost.
6) Environment AI for Smart Wireless: IRS can be used
as one of the following three functions: 1) Passive Relay, 2)
Passive Transmitter, 3) both of them, where the quality of
primary signal is enhanced by passive beamforming via relay
and at the same time, the secondary information generated
from the IRS itself can be embedded in the primary signal
(like ambient backscatter), e.g., [28], [30], [113], [115]. For
example, the IRS may be equipped with sensors monitoring
environments, which generate such secondary information to
be reported to the IoT gateway in the uplink. Therefore,
the mode switching at IRS will need to be intelligently and
remotely performed by the control center through the IoT
gateway, e.g., edge node, considering the user objectives and
device positions. Moreover, if a large number of IRS is
deployed in a certain area to assist the primary transmission
while transmitting their own secondary information (from IoT
sensors), the global control of these IRSs will need to be
handled by the control center by gathering the user objectives
and device positions, so as to assure the optimal routing of
air routes of IRSs in conjunction with the mode switching.
However, due to the latency and privacy issues, the global
control by the control center may not be feasible. Instead, the
collaborative (federated) learning will play a crucial role to
intelligently perform the required global control through the
cooperation with the edge nodes, which perform the learning
locally and upload their model parameters to the control center.
This way we can resolve the latency and privacy issues.
VII. CONCLUSIONS
This paper has presented a comprehensive survey on the
design and applications of the IRS to wireless communication
networks. Firstly, we have presented an overview of the meta-
surface and its reconfigurability to realize the vision of IRS.
Then, we have focused on its applications in wireless networks
and reviewed different network scenarios that can benefit from
its reconfigurability. Afterwards, we have provided detailed
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
reviews on the performance analysis and optimization of IRS-
assisted wireless networks under different communication sce-
narios, including SNR/capacity maximization, transmit power
minimization, EE/SE performance maximization, and secrecy
rate maximization, etc. Due to its diversified applications, we
have also reviewed the emerging use of the IRS to promote
wireless power transfer, UAV communications, and MEC.
Finally, we have outlined important challenges, open issues
as well as future research directions.
This survey covers a broad set of research topics regarding
IRS and its applications in wireless networks, from its physical
characterization and channel modeling, to research problems
in a wireless networking perspective, mainly focusing on the
optimization models and solution methods for IRS-assisted
wireless systems. The current literature has identified ba-
sic problem models and solution methods for IRS-assisted
wireless systems, including tentative discussions on practical
limitations, such as the IRS’s finite phase resolution, partial
channel information, and hardware impairments. In the current
literature, a large majority of the solution methods are based
on alternating optimization framework, which ensures the
convergence to a local-optimal solution. Though significant
performance improvement can be verified by numerical results
and simulations, the performance gap with the optimum is
not known exactly and seldom characterized in literature. In
the future work, by exploring more sophisticated optimization
algorithms, the IRS-assisted wireless systems are expected to
achieve a higher performance gain than that in the current
literature. Besides, the current research works are mainly
limited to joint optimization problems of active and passive
beamforming under different network scenarios. The overall
performance gain of IRS-assisted wireless systems can be
better explored in the future work if the size and distribution
of the IRS’s scattering elements, the orientation and mobility
of the IRS tiles, their partitions and grouping strategies,
etc., are all taken into account, in combination with multi-
users’ access control, user association, information encoding,
transmit scheduling, QoS provisioning, etc. This will open a
significantly larger space to explore than the research scope in
the current literature.
R EFERENCES
[1] S. Zhang, Q. Wu, S. Xu, and G. Y. Li, “Fundamental green tradeoffs:
Progresses, challenges, and impacts on 5G networks,” IEEE Commun.
Surv. Tut., vol. 19, no. 1, pp. 33–56, Firstquarter 2017.
[2] X. Cao, L. Liu, Y. Cheng, and X. S. Shen, “Towards energy-efficient
wireless networking in the big data era: A survey,” IEEE Commun.
Surv. Tut., vol. 20, no. 1, pp. 303–332, Firstquarter 2018.
[3] D. Muirhead, M. A. Imran, and K. Arshad, “A survey of the challenges,
opportunities and use of multiple antennas in current and future 5G
small cell base stations,” IEEE Access, vol. 4, pp. 2952–2964, May
2016.
[4] H. A. U. Mustafa, M. A. Imran, M. Z. Shakir, A. Imran, and
R. Tafazolli, “Separation framework: An enabler for cooperative and
D2D communication for future 5G networks,” IEEE Commun. Surv.
Tut., vol. 18, no. 1, pp. 419–445, Firstquarter 2016.
[5] F. S. Shaikh and R. Wismüller, “Routing in multi-hop cellular device-
to-device (D2D) networks: A survey,” IEEE Commun. Surv. Tut.,
vol. 20, no. 4, pp. 2622–2657, Fourthquarter 2018.
[6] M. Kamel, W. Hamouda, and A. Youssef, “Ultra-dense networks: A
survey,” IEEE Commun. Surv. Tut., vol. 18, no. 4, pp. 2522–2545,
Fourthquarter 2016.
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
27
[7] P. Park, S. Coleri Ergen, C. Fischione, C. Lu, and K. H. Johansson,
“Wireless network design for control systems: A survey,” IEEE Com-
mun. Surv. Tut., vol. 20, no. 2, pp. 978–1013, Secondquarter 2018.
[8] D. Wang, B. Bai, W. Zhao, and Z. Han, “A survey of optimization
approaches for wireless physical layer security,” IEEE Commun. Surv.
Tut., vol. 21, no. 2, pp. 1878–1911, Secondquarter 2019.
[9] L. Christos, N. Shuai, T. Ageliki, P. Andreas, I. Sotiris, and I. F.
Akyildiz, “Realizing wireless communication through software-defined
hypersurface environments,” in proc. IEEE Int. Sym. WoWMoM, Jun.
2018.
[10] M. D. Renzo, M. Debbah, D. T. P. Huy, A. Zappone, M. Aloui-
ni, C. Yuen, V. Sciancalepore, G. C. Alexandropoulos, J. Hoydis,
H. Gacanin, J. de Rosny, A. Bounceu, G. Lerosey, and M. Fink, “Smart
radio environments empowered by AI reconfigurable meta-surfaces: An
idea whose time has come,” EURASIP J. Wireless Commun. Network.,
vol. 129, May 2019.
[11] Q. Wu and R. Zhang, “Towards smart and reconfigurable environment:
Intelligent reflecting surface aided wireless network,” IEEE Commun.
Mag., vol. 58, no. 1, pp. 106–112, Jan. 2020.
[12] E. Basar, M. D. Renzo, J. de Rosny, M. Debbah, M.-S. Alouini, and
R. Zhang, “Wireless communications through reconfigurable intelligent
surfaces,” IEEE Access, vol. 7, pp. 116 753–116 773, 2019.
[13] C. Huang, S. Hu, G. C. Alexandropoulos, A. Zappone, C. Yuen,
R. Zhang, M. D. Renzo, and M. Debbah, “Holographic MIMO
surfaces for 6G wireless networks: Opportunities, challenges, and
trends,” arXiv preprint arXiv:1911.12296, 2019. [Online]. Available:
https://arxiv.org/abs/1911.12296.
[14] Y.-C. Liang, R. Long, Q. Zhang, J. Chen, H. V. Cheng, and H. Guo,
“Large intelligent surface/antennas (LISA): making reflective radios
smart,” Journal of Communications and Information Networks, vol. 4,
no. 2, pp. 40–50, Jun. 2019.
[15] L. Subrt and P. Pechac, “Controlling propagation environments using
intelligent walls,” in proc. European Conf. Antennas Propag. (EUCAP),
Mar. 2012.
[16] L. Subrt and P. Pechac, “Intelligent walls as autonomous parts of smart
indoor environments,” IET Commun., vol. 6, no. 8, pp. 1004–1010, May
2012.
[17] N. Kaina, M. Dupre, G. Lerosey, and M. Fink, “Shaping complex
microwave fields in reverberating media with binary tunable metasur-
faces,” Scientific Reports, vol. 4, no. 6693, Oct. 2014.
[18] A. Welkie, L. Shangguan, J. Gummeson, W. Hu, and K. Jamieson,
“Programmable radio environments for smart spaces,” in proc. ACM
Workshop on Hot Topics in Networks (HotNets). New York, NY, USA:
ACM, Nov. 2017, pp. 36–42.
[19] C. Liaskos, S. Nie, A. Tsioliaridou, A. Pitsillides, S. Ioannidis,
and I. Akyildiz, “A new wireless communication paradigm through
software-controlled metasurfaces,” IEEE Commun. Mag., vol. 56, no. 9,
pp. 162–169, Sept. 2018.
[20] C. Liaskos, S. Nie, A. Tsioliaridou, A. Pitsillides, S. Ioannidis, and
I. F. Akyildiz, “A novel communication paradigm for high capacity
and security via programmable indoor wireless environments in next
generation wireless systems,” Ad Hoc Networks, vol. 87, pp. 1–16, May
2019.
[21] S. Nie, J. M. Jornet, and I. F. Akyildiz, “Intelligent environments
based on ultra-massive MIMO platforms for wireless communication
in millimeter wave and terahertz bands,” in proc. IEEE ICASSP, May
2019, pp. 7849–7853.
[22] M. Najafi and R. Schober, “Intelligent reflecting surfaces for free space
optical communications,” in proc. IEEE GLOBECOM, Dec. 2019, pp.
1–7.
[23] P. del Hougne, M. Fink, and G. Lerosey, “Optimal communication
channels in a disordered world with tamed randomness,” Nature
Electronics, vol. 2, pp. 36–41, Jan. 2019.
[24] Kihun Chang, Sang il Kwak, and Young Joong Yoon, “Equivalent
circuit modeling of active frequency selective surfaces,” in proc. IEEE
Radio Wireless Sym., Jan. 2008, pp. 663–666.
[25] C. Liaskos, A. Tsioliaridou, S. Nie, A. Pitsillides, S. Ioannidis, and
I. F. Akyildiz, “On the network-layer modeling and configuration of
programmable wireless environments,” IEEE/ACM Trans. Network.,
vol. 27, no. 4, pp. 1696–1713, Aug. 2019.
[26] M. D. Alessio Zappone, Marco Di Renzo, “Wireless networks design
in the era of deep learning: Model-based, AI-based, or both?” IEEE
Trans. Commun., vol. 67, no. 10, pp. 7331–7376, Oct. 2019.
[27] N. Van Huynh, D. T. Hoang, X. Lu, D. Niyato, P. Wang, and D. I.
Kim, “Ambient backscatter communications: A contemporary survey,”
IEEE Commun. Surv. Tut., vol. 20, no. 4, pp. 2889–2922, Fourthquarter
2018.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
[28] S. Y. Park and D. I. Kim, “Intelligent reflecting surface-aided phase-
shift backscatter communication,” in proc. Int. Conf. Ubiquit. Inf.
Manag. Commun. (IMCOM), Jan. 2020, pp. 1–5.
[29] Q. Zhang, Y.-C. Liang, and H. V. Poor, “Large intelligent
surface/antennas (LISA) assisted symbiotic radio for IoT
communications,” arXiv preprint arXiv:2002.00340, 2020. [Online].
Available: https://arxiv.org/abs/2002.00340.
[30] H. Zheng, Z. Yang, G. Wang, R. He, and B. Ai, “Channel estimation
for ambient backscatter communications with large intelligent surface,”
in proc. IEEE Int. Conf. Wireless Commun. Signal Process. (WCSP),
Oct. 2019, pp. 1–5.
[31] C. L. Holloway, E. F. Kuester, J. A. Gordon, J. O’Hara, J. Booth, and
D. R. Smith, “An overview of the theory and applications of meta-
surfaces: The two-dimensional equivalents of metamaterials,” IEEE
Antennas Propag. Mag., vol. 54, no. 2, pp. 10–35, Apr. 2012.
[32] H.-T. Chen, A. J. Taylor, and N. Yu, “A review of metasurfaces: physics
and applications,” Reports on Progress in Physics, vol. 79, no. 7, Jun.
2016.
[33] D. Fei, P. Anders, and S. I. Bozhevolnyi, “Gradient metasurfaces:
a review of fundamentals and applications,” Reports on Progress in
Physics, vol. 81, no. 2, Dec. 2017.
[34] N. Bonod, “Large-scale dielectric metasurfaces,” Nature Materials,
vol. 14, pp. 664–665, Jun. 2015.
[35] R. Pestourie, C. Pérez-Arancibia, Z. Lin, W. Shin, F. Capasso, and S. G.
Johnson, “Inverse design of large-area metasurfaces,” Optics Express,
vol. 26, no. 26, pp. 33 732–33 747, Dec. 2018.
[36] Y. Vahabzadeh, N. Chamanara, K. Achouri, and C. Caloz, “Computa-
tional analysis of metasurfaces,” IEEE J. Multiscale and Multiphysics
Computational Techniques, vol. 3, pp. 37–49, Apr. 2018.
[37] W. Kedi, C. Philippe, W. Qi Jie, and G. Patrice, “Modelling of free-form
conformal metasurfaces,” Nature Communications, vol. 9, no. 3494,
Aug. 2018.
[38] K. Achouri, M. A. Salem, and C. Caloz, “General metasurface syn-
thesis based on susceptibility tensors,” IEEE Trans. Antennas Propag.,
vol. 63, no. 7, pp. 2977–2991, Jul. 2015.
[39] G. Lavigne, K. Achouri, V. S. Asadchy, S. A. Tretyakov, and C. Caloz,
“Susceptibility derivation and experimental demonstration of refracting
metasurfaces without spurious diffraction,” IEEE Trans. Antennas
Propag., vol. 66, no. 3, pp. 1321–1330, Mar. 2018.
[40] F. Liu, A. Pitilakis, M. S. Mirmoosa, O. Tsilipakos, X. Wang, A. C.
Tasolamprou, S. Abadal, A. Cabellos-Aparicio, E. Alarcón, C. Liaskos,
N. V. Kantartzis, M. Kafesaki, E. N. Economou, C. M. Soukoulis,
and S. Tretyakov, “Programmable metasurfaces: State of the art and
prospects,” in proc. IEEE Int. Sym. Circuits Syst. (ISCAS), May 2018.
[41] H. Yang, X. Cao, F. Yang, J. Gao, S. Xu, M. Li, X. Chen, Y. Zhao,
Y. Zheng, and L. Sijia, “A programmable metasurface with dynamic
polarization, scattering and focusing control,” Scientific Reports, vol. 6,
no. 35692, Oct. 2016.
[42] C. Liaskos, A. Tsioliaridou, A. Pitsillides, I. F. Akyildiz, N. V.
Kantartzis, A. X. Lalas, X. Dimitropoulos, S. Ioannidis, M. Kafesaki,
and C. M. Soukoulis, “Design and development of software defined
metamaterials for nanonetworks,” IEEE Circuits Syst. Mag., vol. 15,
no. 4, pp. 12–25, Fourthquarter 2015.
[43] N. Kaina, M. Dupré, M. Fink, and G. Lerosey, “Hybridized resonances
to design tunable binary phase metasurface unit cells,” Optics Express,
vol. 22, no. 16, pp. 18 881–1888, Aug. 2014.
[44] J. Zhao, Q. Cheng, J. Chen, M. Q. Qi, W. X. Jiang, and T. J. Cui, “A
tunable metamaterial absorber using varactor diodes,” New Journal of
Physics, vol. 15, no. 4, Apr. 2013.
[45] H. K. Kim, D. Lee, and S. Lim, “Frequency-tunable metamaterial
absorber using a varactor-loaded fishnet-like resonator,” Applied Optics,
vol. 55, no. 15, pp. 4113–4118, May 2016.
[46] F. Liu, O. Tsilipakos, A. Pitilakis, A. C. Tasolamprou, M. S. Mirmoosa,
N. V. Kantartzis, D.-H. Kwon, J. Georgiou, K. Kossifos, M. A. Antoni-
ades, M. Kafesaki, C. M. Soukoulis, and S. A. Tretyakov, “Intelligent
metasurfaces with continuously tunable local surface impedance for
multiple reconfigurable functions,” Physical Review Applied, vol. 11,
pp. 044 024–044 033, Apr. 2019.
[47] S. Abadal, C. Liaskos, A. Tsioliaridou, S. Ioannidis, A. Pitsillides,
J. Solé-Pareta, E. Alarcón, and A. Cabellos-Aparicio, “Computing and
communications for the software-defined metamaterial paradigm: A
context analysis,” IEEE Access, vol. 5, pp. 6225–6235, Apr. 2017.
[48] A. C. Tasolamprou, M. S. Mirmoosa, O. Tsilipakos, A. Pitilakis,
F. Liu, S. Abadal, A. Cabellos-Aparicio, E. Alarcón, C. Liaskos,
N. V. Kantartzis, S. Tretyakov, M. Kafesaki, E. N. Economou, and
C. M. Soukoulis, “Intercell wireless communication in software-defined
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
28
metasurfaces,” in proc. IEEE Int. Sym. Circuits Syst. (ISCAS), May
2018, pp. 1–5.
[49] P. Kouvaros, D. Kouzapas, A. Philippou, J. Georgiou, L. Petrou, and
A. Pitsillides, “Formal verification of a programmable hypersurface,” in
proc. Int. Workshop on Formal Methods for Industrial Critical Systems,
Aug. 2018, pp. 83–97.
[50] T. Saeed, C. Skitsas, D. Kouzapas, M. Lestas, V. Soteriou, A. Philippou,
S. Abadal, C. Liaskos, L. Petrou, J. Georgiou, and A. Pitsillides,
“Fault adaptive routing in metasurface controller networks,” in proc.
Int. Workshop Netw. on Chip Arch. (NoCArc), Oct. 2018, pp. 1–6.
[51] L. Petrou, P. Karousios, and J. Georgiou, “Asynchronous circuits as
an enabler of scalable and programmable metasurfaces,” in proc. IEEE
Int. Sym. Circuits Syst. (ISCAS), May 2018.
[52] S. Abadal, A. Mestres, J. Torrellas, E. Alarcon, and A. Cabellos-
Aparicio, “Medium access control in wireless network-on-chip: A
context analysis,” IEEE Commun. Mag., vol. 56, no. 6, pp. 172–178,
Jun. 2018.
[53] D. S. Dong, J. Yang, Q. Cheng, J. Zhao, L. H. Gao, S. J. Ma, S. Liu,
H. B. Chen, Q. He, W. W. Liu, Z. Fang, L. Zhou, and T. J. Cui,
“Terahertz broadband low-reflection metasurface by controlling phase
distributions,” Advanced Optical Materials, vol. 3, no. 10, pp. 1405–
1410, Jun. 2015.
[54] S. Foo, “Liquid-crystal reconfigurable metasurface reflectors,” in proc.
IEEE Int. Sym. Antennas Propag. USNC/URSI National Radio Science
Meeting, Jul. 2017, pp. 2069–2070.
[55] J. Li, C. Shen, A. Diaz Rubio, S. Tretyakov, and S. Cummer,
“Systematic design and experimental demonstration of bianisotropic
metasurfaces for scattering-free manipulation of acoustic wavefronts,”
Nature Communications, vol. 9, no. 1342, Apr. 2018.
[56] C. Liaskos, A. Tsioliaridou, and S. Ioannidis, “Towards a circular
economy via intelligent metamaterials,” in proc. IEEE Int. Workshop
on Computer Aided Modeling and Design of Communication Links and
Networks (CAMAD), Sept. 2018.
[57] L. Dai, B. Wang, M. Wang, X. Yang, J. Tan, S. Bi, S. Xu, F. Yang,
Z. Chen, M. Di Renzo, C. Chae, and L. Hanzo, “Reconfigurable
intelligent surface-based wireless communications: Antenna design,
prototyping, and experimental results,” IEEE Access, vol. 8, pp. 45 913–
45 923, Mar. 2020.
[58] H. Chu, Q. Li, B. Liu, J. Luo, S. Sun, Z. H. Hang, L. Zhou, and
Y. Lai, “A hybrid invisibility cloak based on integration of transparent
metasurfaces and zero-index materials,” Light, Science & Applications,
vol. 7, no. 50, Aug. 2018.
[59] L. Hsu, T. Lepetit, and B. Kanté, “Extremely thin dielectric metasurface
for carpet cloaking,” Progress In Electromagnetics Research, vol. 152,
pp. 33–40, 2015.
[60] K. M. Kossifos, M. A. Antoniades, J. Georgiou, A. H. Jaafar, and
N. T. Kemp, “An optically-programmable absorbing metasurface,” in
proc. IEEE Int. Sym. Circuits Syst. (ISCAS), May 2018.
[61] N. Yu, P. Genevet, M. A. Kats, F. Aieta, J.-P. Tetienne, F. Capasso, and
Z. Gaburro, “Light propagation with phase discontinuities: Generalized
laws of reflection and refraction,” Science, vol. 334, no. 6054, pp. 333–
337, Oct. 2011.
[62] A. Pors and S. I. Bozhevolnyi, “Plasmonic metasurfaces for efficient
phase control in reflection,” Optics Express, vol. 21, no. 22, pp. 27 438–
27 451, 2013.
[63] J. Park, J.-H. Kang, S. J. Kim, X. Liu, and M. L. Brongersma,
“Dynamic reflection phase and polarization control in metasurfaces,”
Nano Letters, vol. 17, no. 1, pp. 407–413, Dec. 2016.
[64] Y. Li, B. Liang, Z. ming Gu, X. ye Zou, and J. chun Cheng,
“Reflected wavefront manipulation based on ultrathin planar acoustic
metasurfaces,” Scientific Reports, vol. 3, no. 2546, Aug. 2013.
[65] A. Díaz-Rubio, V. Asadchy, A. Elsakka, and S. Tretyakov, “Metasur-
faces for perfect control of reflection,” in proc. IEEE Int. Workshop
Antenna Technology: Small Antennas, Innovative Structures, and Ap-
plications (iWAT), Mar. 2017, pp. 3–5.
[66] D.-H. Kwon and S. A. Tretyakov, “Arbitrary beam control using passive
lossless metasurfaces enabled by orthogonally polarized custom surface
waves,” Physical Review B, vol. 97, Oct. 2017.
[67] E. Basar, “Index modulation techniques for 5G wireless networks,”
IEEE Commun. Mag., vol. 54, no. 7, pp. 168–175, Jul. 2016.
[68] Y. Zhou, G. Zhang, H. Chen, P. Zhou, X. Wang, L. Zhang, l. Zhang,
J. Xie, and L. Deng, “Design of phase gradient coding metasurfaces
for broadband wave modulating,” Scientific Reports, vol. 8, no. 8672,
Jun. 2018.
[69] Z. Lei, C. Xiao Qing, L. Shuo, Z. Qian, Z. Jie, D. Jun Yan,
B. Guo Dong, W. Xiang, C. Qiang, C. Giuseppe, G. Vincenzo, and
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
C. Tie Jun, “Space-time-coding digital metasurfaces,” Nature Commu-
nications, vol. 9, no. 4334, Oct. 2018.
[70] A. S. da Silva, F. Monticone, G. Castaldi, V. Galdi, A. Alú, and
N. Engheta, “Performing mathematical operations with metamaterials,”
Science, vol. 343, pp. 160–163, Jan. 2014.
[71] F. Zangeneh-Nejad and R. Fleury, “Performing mathematical op-
erations using high-index acoustic metamaterials,” New Journal of
Physics, vol. 20, no. 7, p. 073001, Jul. 2018.
[72] P. del Hougne and G. Lerosey, “Leveraging chaos for wave-based ana-
log computation: Demonstration with indoor wireless communication
signals,” Physical Review X, vol. 8, Nov. 2018.
[73] H. Tao, L. R. Chieffo, M. A. Brenckle, S. M. Siebert, M. Liu, A. C.
Strikwerda, K. Fan, D. L. Kaplan, X. Zhang, R. D. Averitt, and F. G.
Omenetto, “Metamaterials on paper as a sensing platform,” Advanced
Materials, vol. 23, no. 28, pp. 3197–3201, Jun. 2011.
[74] T. Nakanishi, T. Otani, Y. Tamayama, and M. Kitano, “Storage of elec-
tromagnetic waves in a metamaterial that mimics electromagnetically
induced transparency,” Physical Review B, vol. 87, no. 16, Apr. 2013.
[75] Z. Bai, D. Xu, and G. Huang, “Storage and retrieval of electromagnetic
waves with orbital angular momentum via plasmon-induced transparen-
cy,” Optics Express, vol. 25, no. 2, pp. 785–798, Jan 2017.
[76] C. Huang, C. Zhang, J. Yang, B. Sun, B. Zhao, and X. Luo, “Recon-
figurable metasurface for multifunctional control of electromagnetic
waves,” Advanced Optical Materials, vol. 5, no. 22, Sept. 2017.
[77] K. Bi, D. Yang, J. Chen, Q. Wang, H. Wu, C. Lan, and Y. Yang,
“Experimental demonstration of ultra-large-scale terahertz all-dielectric
metamaterials,” Photonics Research, vol. 7, no. 4, pp. 457–463, Apr.
2019.
[78] E. Carrasco, M. Barba, and J. A. Encinar, “X-band reflectarray antenna
with switching-beam using pin diodes and gathered elements,” IEEE
Trans. Antennas Propag., vol. 60, no. 12, pp. 5700–5708, Dec. 2012.
[79] J. Y. Lau and S. V. Hum, “Reconfigurable transmitarray design
approaches for beamforming applications,” IEEE Trans. Antennas
Propag., vol. 60, no. 12, pp. 5679–5689, Dec. 2012.
[80] B. Zhu, Y. Feng, J. Zhao, C. Huang, and T. Jiang, “Switchable
metamaterial reflector/absorber for different polarized electromagnetic
waves,” Applied Physics Letters, vol. 97, no. 5, p. 051906, Aug. 2010.
[81] X. Tan, Z. Sun, D. Koutsonikolas, and J. Jornet, “Enabling indoor
mobile millimeter-wave networks based on smart reflect-arrays,” in
proc. IEEE INFOCOM, Apr. 2018, pp. 270–278.
[82] N. Dabidian, S. Dutta-Gupta, I. N. Kholmanov, K. Lai, F. Lu, J. Lee,
M. Jin, S. T. Trendafilov, A. B. Khanikaev, B. Fallahazad, E. Tutuc,
M. Belkin, and G. Shvets, “Experimental demonstration of phase mod-
ulation and motion sensing using graphene-integrated metasurfaces.”
Nano Letters, vol. 16, pp. 3607–3615, Jun. 2016.
[83] A. Ourir, K. Rachedi, D. T. Phan-Huy, C. Leray, and J. de Rosny,
“Compact reconfigurable antenna with radiation pattern diversity for
spatial modulation,” in proc. European Conf. Antennas Propag. (EU-
CAP), Mar. 2017, pp. 3038–3043.
[84] Horizon 2020 Program (2014-2020) VISORSURF project, “A
hardware platform for software-driven functional metasurfaces.”
[Online]. Available: http://www.visorsurf.eu/
[85] H. Lu, Y. Zeng, S. Jin, and R. Zhang, “Enabling panoramic full-angle
reflection via aerial intelligent reflecting surface,” in proc. IEEE ICC,
Jun. 2020, pp. 1–6.
[86] W. Tang, M. Z. Chen, X. Chen, J. Y. Dai, Y. Han, M. D. Renzo,
Y. Zeng, S. Jin, Q. Cheng, and T. J. Cui, “Wireless communications
with reconfigurable intelligent surface: Path loss modeling and
experimental measurement,” arXiv preprint arXiv:1911.05326, 2019.
[Online]. Available: https://arxiv.org/abs/1911.05326
[87] O. Özdogan, E. Björnson, and E. G. Larsson, “Intelligent reflecting
surfaces: Physics, propagation, and pathloss modeling,” IEEE Wireless
Commun. Lett., vol. 9, no. 5, pp. 581–585, May 2020.
[88] M. D. Renzo and J. Song, “Reflection probability in wireless networks
with metasurface-coated environmental objects: An approach based on
random spatial processes,” EURASIP J. Wireless Commun. Network.,
vol. 99, Apr. 2019.
[89] T. Bai, R. Vaze, and R. W. H. Jr., “Analysis of blockage effects on urban
cellular networks,” IEEE Trans. Wireless Commun., vol. 13, no. 9, pp.
5070–5083, Sep. 2014.
[90] C. Guo, Y. Cui, F. Yang, and L. Ding, “Outage probability analysis and
minimization in intelligent reflecting surface-assisted MISO systems,”
IEEE Commun. Lett., pp. 1–1, Feb. 2020.
[91] A. Narayanan, S. T. V., and R. K. Ganti, “Coverage analysis in
millimeter wave cellular networks with reflections,” in proc. IEEE
GLOBECOM, Dec. 2017.
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
29
[92] D. Nguyen Viet, M. Di Renzo, V. Basavarajappa, B. Bedia Exposito,
J. Basterrechea, and D.-T. Phan-Huy, “Spatial modulation based on
reconfigurable antennas: Performance evaluation by using the prototype
of a reconfigurable antenna,” EURASIP J. Wireless Commun. Network.,
vol. 149, Jun. 2019.
[93] V. C. Thirumavalavan and T. S. Jayaraman, “BER analysis of recon-
figurable intelligent surface assisted downlink power domain NOMA
system,” in proc. IEEE Int. Conf. COmmun. Syst. Netw. (COMSNETS),
Jan. 2020, pp. 519–522.
[94] E. Basar, “Transmission through large intelligent surfaces: A new fron-
tier in wireless communications,” in proc. European Conf. Networks
and Commun. (EuCNC), Jun. 2019, pp. 112–117.
[95] M. Jung, W. Saad, M. Debbah, and C. S. Hong, “Asymptotic opti-
mality of reconfigurable intelligent surfaces: Passive beamforming and
achievable rate,” in proc. IEEE ICC, Jun. 2020.
[96] J. He, H. Wymeersch, T. Sanguanpuak, O. Silvén, and M. Juntti, “Adap-
tive beamforming design for mmWave RIS-aided joint localization and
communication,” in proc. IEEE WCNC, Apr. 2020.
[97] Özgecan Özdogan, E. Björnson, and E. G. Larsson, “Using intelligent
reflecting surfaces for rank improvement in MIMO communications,”
in proc. IEEE ICASSP, May 2020.
[98] H. Zhang, B. Di, L. Song, and Z. Han, “Reconfigurable intelligent
surfaces assisted communications with limited phase shifts: How many
phase shifts are enough?” IEEE Trans. Veh. Techn., vol. 69, no. 4, pp.
4498–4502, Apr. 2020.
[99] J. V. Alegría and F. Rusek, “Achievable rate with correlated hardware
impairments in large intelligent surfaces,” in proc. IEEE Int. Workshop
Computat. Adv. Multi-Sensor Adaptive Process. (CAMSAP), Dec. 2019,
pp. 559–563.
[100] J. Lyu and R. Zhang, “Spatial throughput characterization for intelligent
reflecting surface aided multiuser system,” IEEE Wireless Commun.
Lett., vol. 9, no. 6, pp. 834–838, Jun. 2020.
[101] S. Zhang and R. Zhang, “Intelligent reflecting surface aided multiple
access: capacity region and deployment strategy,” in proc. IEEE Int.
Workshop Signal Process. Adv. Wireless Commun. (SPAWC), Feb. 2020.
[102] X. Tan, Z. Sun, J. M. Jornet, and D. Pados, “Increasing indoor spectrum
sharing capacity using smart reflect-array,” in proc. IEEE ICC, May
2016.
[103] H. Sha, R. Fredrik, and E. Ove, “Beyond massive-MIMO: The potential
of data-transmission with large intelligent surfaces,” IEEE Trans. Signal
Process., vol. 66, no. 10, pp. 2746–2758, May 2018.
[104] J. V. Alegria, J. Rodriguez Sanchez, F. Rusek, L. Liu, and O. Edfors,
“Decentralized equalizer construction for large intelligent surfaces,” in
proc. IEEE VTC (Fall), Sep. 2019, pp. 1–6.
[105] M. Jung, W. Saad, Y. Jang, G. Kong, and S. Choi, “Performance
analysis of large intelligent surfaces (LISs): Asymptotic data rate and
channel hardening effects,” IEEE Trans. Wireless Commun., vol. 19,
no. 3, pp. 2052–2065, Mar. 2020.
[106] M. Jung, W. Saad, Y. Jang, G. Kong, and S. Choi, “Reliability
analysis of large intelligent surfaces (LISs): Rate distribution and
outage probability,” IEEE Wireless Commun. Lett., vol. 8, no. 6, pp.
1662–1666, Dec. 2019.
[107] H. Sha, R. Fredrik, and E. Ove, “Capacity degradation with modeling
hardware impairment in large intelligent surface,” in proc. IEEE
GLOBECOM, Dec. 2018.
[108] S. Hu, F. Rusek, and O. Edfors, “Beyond massive MIMO: The potential
of positioning with large intelligent surfaces,” IEEE Trans. Signal
Process., vol. 66, no. 7, pp. 1761–1774, Apr. 2018.
[109] E. Basar, M. Wen, R. Mesleh, M. Di Renzo, Y. Xiao, and H. Haas,
“Index modulation techniques for next-generation wireless networks,”
IEEE Access, vol. 5, pp. 16 693–16 746, Aug. 2017.
[110] M. Di Renzo, “Spatial modulation based on reconfigurable antennas -
a new air interface for the IoT,” in proc. IEEE MILCOM, Oct. 2017,
pp. 495–500.
[111] D. T. P. Huy, Y. Kokar, K. Rachedi, P. Pajusco, A. Mokh, T. Magounaki,
R. Masood, C. Buey, P. Ratajczak, N. Malhouroux-Gaffet, J. Conrat,
J. Prévotet, A. Ourir, J. de Rosny, M. Crussiére, M. Hélard, A. Gati,
T. Sarrebourse, and M. D. Renzo, “Single-carrier spatial modulation
for the internet of things: Design and performance evaluation by using
real compact and reconfigurable antennas,” IEEE Access, vol. 7, pp.
18 978–18 993, Jan. 2019.
[112] E. Basar, “Reconfigurable intelligent surface-based index modulation:
A new beyond MIMO paradigm for 6G,” IEEE Trans. Commun.,
vol. 68, no. 5, pp. 3187–3196, May 2020.
[113] W. Yan, X. Yuan, and X. Kuai, “Passive beamforming and information
transfer via large intelligent surface,” IEEE Wireless Commun. Lett.,
vol. 9, no. 4, pp. 533–537, Apr. 2020.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
[114] W. Zhao, G. Wang, S. Atapattu, T. A. Tsiftsis, and C. Tellambura, “Is
backscatter link stronger than direct link in reconfigurable intelligent
surface-assisted system?” IEEE Commun. Lett., vol. 24, no. 6, pp.
1342–1346, Jun. 2020.
[115] X. Guan, Q. Wu, and R. Zhang, “Joint power control and passive
beamforming in IRS-assisted spectrum sharing,” IEEE Commun. Lett.,
pp. 1–1, Mar. 2020.
[116] Q. Wu and R. Zhang, “Intelligent reflecting surface enhanced wireless
network: Joint active and passive beamforming design,” in proc. IEEE
GLOBECOM, Dec. 2018, pp. 1–6.
[117] E. Björnson and L. Sanguinetti, “Demystifying the power scaling law
of intelligent reflecting surfaces and metasurfaces,” in proc. IEEE Int.
Workshop Computat. Adv. Multi-Sensor Adaptive Process. (CAMSAP),
Dec. 2019.
[118] X. Yu, D. Xu, and R. Schober, “MISO wireless communication systems
via intelligent reflecting surfaces,” in proc. IEEE/CIC ICCC, Aug.
2019, pp. 735–740.
[119] Y. Yang, S. Zhang, and R. Zhang, “IRS-enhanced OFDM: Power
allocation and passive array optimization,” in proc. IEEE GLOBECOM,
Dec. 2019, pp. 1–6.
[120] J. Yuan, Y.-C. Liang, J. J. G. Feng, and E. G. Larsson,
“Intelligent reflecting surface-assisted cognitive radio system,”
arXiv preprint arXiv:1912.10678, 2020. [Online]. Available:
https://arxiv.org/abs/1912.10678.
[121] P. Wang, J. Fang, X. Yuan, Z. D. Chen, H. Duan, and H. Li, “Intelligent
reflecting surface-assisted millimeter wave communications: Joint
active and passive precoding design,” arXiv preprint arXiv:1908.10734,
2019. [Online]. Available: https://arxiv.org/abs/1908.10734
[122] S. Zhang and R. Zhang, “Capacity characterization for intelligent
reflecting surface aided MIMO communication,” IEEE J. Sel. Area.
Commun., Jun. 2020.
[123] K. Ying, Z. Gao, S. Lyu, Y. Wu, H. Wang, and M. Alouini, “GMD-
based hybrid beamforming for large reconfigurable intelligent surface
assisted millimeter-wave massive MIMO,” IEEE Access, vol. 8, pp.
19 530–19 539, Jan. 2020.
[124] C. Guo, Z. Lu, Z. Guo, F. Yang, and L. Ding, “Maximum ergod-
ic capacity of intelligent reflecting surface assisted MIMO wireless
communication system,” in proc. Int. Conf. Commun. Network. China.
Springer International Publishing, Feb. 2020, pp. 331–343.
[125] B. Ning, Z. Chen, W. Chen, and J. Fang, “Beamforming optimiza-
tion for intelligent reflecting surface assisted mimo: A sum-path-gain
maximization approach,” IEEE Wireless Commun. Lett., pp. 1–1, Mar.
2020.
[126] Y. Zhang, C. Zhong, Z. Zhang, and W. Lu, “Sum rate optimization
for two way communications with intelligent reflecting surface,” IEEE
Commun. Lett., vol. 24, no. 4, pp. 1090–1094, May 2020.
[127] S. Abeywickrama, R. Zhang, and C. Yuen, “Intelligent reflecting
surface: Practical phase shift model and beamforming optimization,”
IEEE Trans. Commun., Jun. 2020.
[128] Y. Han, W. Tang, S. Jin, C. Wen, and X. Ma, “Large intelligent
surface-assisted wireless communication exploiting statistical CSI,”
IEEE Trans. Veh. Techn., vol. 68, no. 8, pp. 8238–8242, Aug. 2019.
[129] C. You, B. Zheng, and R. Zhang, “Progressive channel estimation and
passive beamforming for intelligent reflecting surface with discrete
phase shifts,” arXiv preprint arXiv:1912.10646, 2020. [Online].
Available: https://arxiv.org/abs/1912.10646.
[130] J. Xu, W. Xu, and A. L. Swindlehurst, “Discrete phase shift design
for practical large intelligent surface communication,” in proc. IEEE
Pacific Rim Conf. Commun., Comput. Signal Process. (PACRIM), Aug.
2019, pp. 1–5.
[131] Y. Yang, B. Zheng, S. Zhang, and R. Zhang, “Intelligent reflecting
surface meets OFDM: Protocol design and rate maximization,” IEEE
Trans. Commun., Mar. 2020.
[132] M. Badiu and J. P. Coon, “Communication through a large reflecting
surface with phase errors,” IEEE Wireless Commun. Lett., vol. 9, no. 2,
pp. 184–188, Feb. 2020.
[133] M. Jung, W. Saad, and G. Kong, “Spectral efficiency in large intelligent
surfaces: Asymptotic analysis under pilot contamination,” in proc.
IEEE GLOBECOM, Dec. 2019, pp. 1–6.
[134] C. Huang, A. Zappone, M. Debbah, and C. Yuen, “Achievable rate
maximization by passive intelligent mirrors,” in proc. IEEE ICASSP,
Apr. 2018, pp. 3714–3718.
[135] Y. Gao, C. Yong, Z. Xiong, D. Niyato, Y. Xiao, and J. Zhao, “Recon-
figurable intelligent surface for MISO systems with proportional rate
constraints,” in proc. IEEE ICC, Jun. 2020, pp. 1–6.
[136] H. Guo, Y. Liang, J. Chen, and E. G. Larsson, “Weighted sum-
rate maximization for reconfigurable intelligent surface aided wireless
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
30
networks,” IEEE Trans. Wireless Commun., vol. 19, no. 5, pp. 3064–
3076, Feb. 2020.
[137] M. M. Zhao, Q. Wu, M. J. Zhao, and R. Zhang, “Two-timescale
beamforming optimization for intelligent reflecting surface enhanced
wireless network,” in proc. IEEE Sensor Array and Multichannel Signal
Process. Workshop, Jun. 2020.
[138] Z. Ding and H. V. Poor, “A simple design of IRS-NOMA transmission,”
IEEE Commun. Lett., vol. 24, no. 5, pp. 1119–1123, May 2020.
[139] Y. Liu, L. Zhang, B. Yang, W. Guo, and M. Imran, “Programmable
wireless channel for multi-user MIMO transmission using meta-
surface,” in Proc. IEEE GLOBECOM 2019, Dec. 2019.
[140] G. Zhou, C. Pan, H. Ren, K. Wang, and A. Nallanathan, “Intelligent
reflecting surface aided multigroup multicast MISO communication
systems,” IEEE Trans. Signal Process., vol. 68, pp. 3236–3251, Apr.
2020.
[141] C. Pan, H. Ren, K. Wang, W. Xu, M. Elkashlan, A. Nallanathan, and
L. H. Hanzo, “Multicell MIMO communications relying on intelligent
reflecting surface,” IEEE Trans. Wireless Commun., May 2020.
[142] G. Yang, X. Xu, and Y.-C. Liang, “Intelligent reflecting surface assisted
non-orthogonal multiple access,” in proc. IEEE WCNC, Apr. 2020.
[143] H. Sha, C. Krishna, R. Fredrik, and E. Ove, “User assignment with
distributed large intelligent surface (LIS) systems,” in proc. IEEE
PIMRC, Bologna, Italy, Sept. 2018.
[144] H. Xie, J. Xu, and Y.-F. Liu, “Max-min fairness in IRS-aided multi-cell
MISO systems via joint transmit and reflective beamforming,” in proc.
IEEE ICC, Jun. 2020.
[145] R. Liu, H. Li, M. Li, and Q. Liu, “Symbol-level precoding design
for intelligent reflecting surface assisted multi-user mimo systems,” in
proc. Int. Conf. Wireless Commun. Signal Process. (WCSP), Oct. 2019,
pp. 1–6.
[146] H. Guo, Y. Liang, J. Chen, and E. G. Larsson, “Weighted sum-
rate maximization for intelligent reflecting surface enhanced wireless
networks,” in proc. IEEE GLOBECOM, Dec. 2019, pp. 1–6.
[147] B. Di, H. Zhang, L. Song, Y. Li, and Z. Han, “Practical hybrid
beamforming with finite-resolution phase shifters for reconfigurable
intelligent surface based multi-user communications,” IEEE Trans. Veh.
Techn., vol. 69, no. 4, pp. 4565–4570, Apr. 2020.
[148] Y. Yang, S. Zhang, and R. Zhang, “IRS-enhanced OFDMA: Joint
resource allocation and passive beamforming optimization,” IEEE
Wireless Commun. Lett., vol. 9, no. 6, pp. 760–764, Jun. 2020.
[149] Q. Wu and R. Zhang, “Intelligent reflecting surface enhanced wireless
network via joint active and passive beamforming,” IEEE Trans.
Wireless Commun., vol. 18, no. 11, pp. 5394–5409, Nov. 2019.
[150] H. Han, J. Zhao, D. Niyato, M. D. Renzo, and Q.-V. Pham, “Intelligent
reflecting surface aided network: Power control for physical-layer
broadcasting,” in proc. IEEE ICC, Jun. 2020.
[151] M. Fu, Y. Zhou, and Y. Shi, “Intelligent reflecting surface for downlink
non-orthogonal multiple access networks,” in proc. IEEE GLOBECOM
Workshops, Dec. 2019.
[152] Y. Li, M. Jiang, Q. Zhang, and J. Qin, “Joint beamforming design
in multi-cluster MISO NOMA intelligent reflecting surface-aided
downlink communication networks,” arXiv preprint arXiv:1909.06972.
[Online]. Available: https://arxiv.org/abs/1909.06972
[153] B. Zheng, Q. Wu, and R. Zhang, “Intelligent reflecting surface-assisted
multiple access with user pairing: NOMA or OMA?” IEEE Commun.
Lett., vol. 24, no. 4, pp. 753–757, Apr. 2020.
[154] C. Huang, A. Zappone, G. C. Alexandropoulos, M. Debbah, and
C. Yuen, “Reconfigurable intelligent surfaces for energy efficiency in
wireless communication,” IEEE Trans. Wireless Commun., vol. 18,
no. 8, Jun. 2019.
[155] E. Björnson, Ö. Özdogan, and E. G. Larsson, “Intelligent reflecting
surface versus decode-and-forward: How large surfaces are needed to
beat relaying?” IEEE Wireless Commun. Lett., vol. 9, pp. 244–248,
Feb. 2020.
[156] Q. Wu and R. Zhang, “Beamforming optimization for intelligent re-
flecting surface with discrete phase shifts,” in proc. IEEE International
Conference on Acoustics, Speech and Signal Processing (ICASSP),
May 2019, pp. 7830–7833.
[157] C. Huang, G. C. Alexandropoulos, A. Zappone, M. Debbah, and
C. Yuen, “Energy efficient multi-user MISO communication using
low resolution large intelligent surfaces,” in proc. IEEE GLOBECOM
(Wkshps), Dec. 2018, pp. 1–6.
[158] Q. Wu and R. Zhang, “Beamforming optimization for wireless network
aided by intelligent reflecting surface with discrete phase shifts,” IEEE
Trans. Commun., vol. 68, no. 3, pp. 1838–1851, Dec. 2020.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
[159] Y. Zou, S. Gong, J. Xu, W. Cheng, D. T. Hoang, and D. Niyato, “Joint
energy beamforming and optimization for intelligent reflecting surface
enhanced communications,” in proc. IEEE WCNC Workshops, Apr.
2020, pp. 1–6.
[160] X. Yu, D. Xu, and R. Schober, “Enabling secure wireless communica-
tions via intelligent reflecting surfaces,” in proc. IEEE GLOBECOM,
Dec. 2019.
[161] O. F and H. B., “The secrecy capacity of the MIMO wiretap channel,”
IEEE Trans. Inf. Theory, vol. 57, no. 8, pp. 4961–4972, Jul. 2011.
[162] H. Shen, W. Xu, S. Gong, Z. He, and C. Zhao, “Secrecy rate
maximization for intelligent reflecting surface assisted multi-antenna
communications,” IEEE Commun. Lett., vol. 23, no. 9, pp. 1488–1492,
Sept. 2019.
[163] L. Dong and H. Wang, “Secure MIMO transmission via intelligent
reflecting surface,” IEEE Wireless Commun. Lett., vol. 9, no. 6, pp.
787–790, Jun. 2020.
[164] B. Feng, Y. Wu, and M. Zheng, “Secure transmission strategy for
intelligent reflecting surface enhanced wireless system,” in proc. IEEE
Int. Conf. Wireless Commun. Signal Process. (WCSP), Oct. 2019, pp.
1–6.
[165] B. Ning, Z. Chen, W. Chen, and L. Li, “Improving security of
THz communication with intelligent reflecting surface,” in proc. IEEE
GLOBECOM Workshops, Dec. 2019, pp. 1–6.
[166] Z. Chu, W. Hao, P. Xiao, and J. Shi, “Intelligent reflecting surface
aided multi-antenna secure transmission,” IEEE Wireless Commun.
Lett., vol. 9, no. 1, pp. 108–112, Jan. 2020.
[167] M. Cui, G. Zhang, and R. Zhang, “Secure wireless communication via
intelligent reflecting surface,” IEEE Wireless Commun. Lett., vol. 8, pp.
1410–1414, Oct. 2019.
[168] J. Chen, Y.-C. Liang, Y. Pei, and H. Guo, “Intelligent reflecting surface:
A programmable wireless environment for physical layer security,”
IEEE Access, vol. 7, pp. 82 599–82 612, Jun. 2019.
[169] X. Guan, Q. Wu, and R. Zhang, “Intelligent reflecting surface assisted
secrecy communication: is artificial noise helpful or not?” IEEE
Wireless Commun. Lett., vol. 9, no. 6, pp. 778–782, Jun. 2020.
[170] D. Xu, X. Yu, Y. Sun, D. W. K. Ng, and R. Schober, “Resource
allocation for secure IRS-assisted multiuser miso systems,” in proc.
IEEE GLOBECOM Workshops, Dec. 2019, pp. 1–6.
[171] Q. Wang, F. Zhou, and R. Q. Hu, “Energy-efficient beamforming and
cooperative jamming in IRS-assisted MISO networks,” in proc. IEEE
ICC, Jun. 2019.
[172] Z. Liu, J. Liu, Y. Zeng, and J. Ma, “Covert wireless communications
in iot systems: Hiding information in interference,” IEEE Wireless
Commun., vol. 25, no. 6, pp. 46–52, Dec. 2018.
[173] X. Lu, E. Hossain, T. Shafique, S. Feng, H. Jiang, and D. Niyato,
“Intelligent reflecting surface (IRS)-enabled covert communications in
wireless networks,” IEEE Network, pp. 1–8, Jun. 2020.
[174] C. Huang, G. C. Alexandropoulos, C. Yuen, and M. Debbah, “Indoor
signal focusing with deep learning designed reconfigurable intelligent
surfaces,” in proc. IEEE SPAWC, Jul. 2019, pp. 1–5.
[175] J. Gao, C. Zhong, X. Chen, H. Lin, and Z. Zhang, “Unsupervised
learning for passive beamforming,” IEEE Commun. Lett., vol. 24, no. 5,
pp. 1052–1056, May 2020.
[176] K. Feng, Q. Wang, X. Li, and C. Wen, “Deep reinforcement learning
based intelligent reflecting surface optimization for MISO communica-
tion systems,” IEEE Wireless Commun. Lett., vol. 9, no. 5, pp. 745–749,
May 2020.
[177] N. C. Luong, D. T. Hoang, S. Gong, D. Niyato, P. Wang, Y. Liang, and
D. I. Kim, “Applications of deep reinforcement learning in communi-
cations and networking: A survey,” IEEE Commun. Surv. Tut., vol. 21,
no. 4, pp. 3133–3174, Fourthquarter 2019.
[178] A. Taha, M. Alrabeiah, and A. Alkhateeb, “Deep learning for large
intelligent surfaces in millimeter wave and massive MIMO systems,”
in proc. GLOBECOM, Dec. 2019, pp. 1–6.
[179] C. Liaskos, A. Tsioliaridou, S. Nie, A. Pitsillides, S. Ioannidis, and
I. Akyildiz, “An interpretable neural network for configuring pro-
grammable wireless environments,” in proc. IEEE Int. Workshop Signal
Process. Adv. Wireless Commun. (SPAWC), Jul. 2019, pp. 1–5.
[180] D. Mishra and H. Johansson, “Channel estimation and low-complexity
beamforming design for passive intelligent surface assisted MISO
wireless energy transfer,” in proc. IEEE ICASSP, May 2019, pp. 4659–
4663.
[181] Q. Wu and R. Zhang, “Weighted sum power maximization for intelli-
gent reflecting surface aided SWIPT,” IEEE Wireless Commun. Lett.,
vol. 9, no. 5, pp. 586–590, May 2020.
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
31
[182] Y. Tang, G. Ma, H. Xie, J. Xu, and X. Han, “Joint transmit and re-
flective beamforming design for IRS-assisted multiuser MISO SWIPT
systems,” in proc. IEEE ICC, Jun. 2020.
[183] C. Pan, H. Ren, K. Wang, M. Elkashlan, A. Nallanathan, J. Wang, and
L. H. Hanzo, “Intelligent reflecting surface aided MIMO broadcasting
for simultaneous wireless information and power transfer,” IEEE J. Sel.
Area. Commun., Jun. 2020.
[184] Y. Zheng, S. Bi, Y. J. Zhang, Z. Quan, and H. Wang, “Intelligent
reflecting surface enhanced user cooperation in wireless powered
communication networks,” IEEE Wireless Commun. Lett., vol. 9, no. 6,
pp. 901–905, Jun. 2020.
[185] B. Lyu, D. T. Hoang, S. Gong, and Z. Yang, “Intelligent reflecting
surface assisted wireless powered communication networks,” in proc.
IEEE WCNC Workshops, Apr. 2020, pp. 1–7.
[186] Q. Wu and R. Zhang, “Joint active and passive beamforming
optimization for intelligent reflecting surface assisted SWIPT under
QoS constraints,” arXiv preprint arXiv:1910.06220, 2019. [Online].
Available: https://arxiv.org/abs/1910.06220
[187] S. Li, B. Duo, X. Yuan, Y. Liang, and M. Di Renzo, “Reconfigurable
intelligent surface assisted UAV communication: Joint trajectory design
and passive beamforming,” IEEE Wireless Commun. Lett., vol. 9, no. 5,
pp. 716–720, May 2020.
[188] D. Ma, M. Ding, and M. Hassan, “Enhancing cellular communications
for UAVs via intelligent reflective surface,” in proc. IEEE WCNC, Apr.
2020.
[189] Q. Zhang, W. Saad, and M. Bennis, “Reflections in the sky: Millimeter
wave communication with UAV-carried intelligent reflectors,” in proc.
IEEE GLOBECOM, Dec. 2019.
[190] T. Jiang and Y. Shi, “Over-the-air computation via intelligent reflecting
surfaces,” in proc. IEEE GLOBECOM, Dec. 2019, pp. 1–6.
[191] S. Hua and Y. Shi, “Reconfigurable intelligent surface for green edge
inference in machine learning,” in proc. IEEE GLOBECOM Workshops,
Dec. 2019, pp. 1–6.
[192] T. L. Jensen and E. D. Carvalho, “An optimal channel estimation
scheme for intelligent reflecting surfaces based on a minimum variance
unbiased estimator,” in proc. IEEE ICASSP, May 2020.
[193] Z. Wang, L. Liu, and S. Cui, “Channel estimation for intelligent
reflecting surface assisted multiuser communications,” in proc. IEEE
WCNC, Apr. 2020.
[194] B. Zheng and R. Zhang, “Intelligent reflecting surface-enhanced OFD-
M: Channel estimation and reflection optimization,” IEEE Wireless
Commun. Lett., vol. 9, no. 4, pp. 518–522, Apr. 2020.
[195] S. Xia and Y. Shi, “Intelligent reflecting surface for massive device
connectivity: Joint activity detection and channel estimation,” in proc.
IEEE ICASSP, May 2020.
[196] B. Zheng, C. You, and R. Zhang, “Intelligent reflecting surface
assisted multi-user OFDMA: channel estimation and training
design,” arXiv preprint arXiv:2003.00648, 2020. [Online]. Available:
https://arxiv.org/abs/2003.00648.
[197] Z. He and X. Yuan, “Cascaded channel estimation for large intelligent
metasurface assisted massive MIMO,” IEEE Wireless Commun. Lett.,
vol. 9, no. 2, pp. 210–214, Feb. 2020.
[198] A. Zappone, M. D. Renzo, M. Debbah, T. L. Thanh, and X. Qian,
“Model-aided wireless artificial intelligence: Embedding expert knowl-
edge in deep neural networks towards wireless systems optimization,”
IEEE Veh. Techn. Mag., vol. 14, no. 3, pp. 60–69, Jul. 2019.
[199] C. Zhang, P. Patras, and H. Haddadi, “Deep learning in mobile and
wireless networking: A survey,” IEEE Commun. Surv. Tut., vol. 21,
no. 3, pp. 2224–2287, Thirdquarter 2019.
[200] N. Van Huynh, D. T. Hoang, X. Lu, D. Niyato, P. Wang, and D. I.
Kim, “Ambient backscatter communications: A contemporary survey,”
IEEE Commun. Surv. Tut., vol. 20, no. 4, pp. 2889–2922, Fourthquarter
2018.
[201] X. Cheng, L. Fang, L. Yang, and S. Cui, “Mobile big data: The fuel for
data-driven wireless,” IEEE Internet of Things Journal, vol. 4, no. 5,
pp. 1489–1516, Oct. 2017.
[202] M. Zhao, Y. Tian, H. Zhao, M. A. Alsheikh, T. Li, R. Hristov,
Z. Kabelac, D. Katabi, and A. Torralba, “RF-based 3D skeletons,” in
proc. ACM SIGCOMM, New York, NY, USA, Aug. 2018, pp. 267–281.
[203] Q. Yan, W. Chen, and H. V. Poor, “Big data driven wireless com-
munications: A human-in-the-loop pushing technique for 5G systems,”
IEEE Wireless Commun., vol. 25, no. 1, pp. 64–69, Feb. 2018.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials
Shimin Gong (M’15) received the B.E. and M.E.
degrees in Electronics and Information Engineering
from Huazhong University of Science and Tech-
nology, Wuhan 430074, China, in 2008 and 2012,
respectively, and the Ph.D. degree in Computer En-
gineering from Nanyang Technological University,
Singapore, in 2014. He was an associate researcher
with the Shenzhen Institutes of Advanced Tech-
nology, Chinese Academy of Sciences, Shenzhen,
China. He is currently an Associate Professor with
the School of Intelligent Systems Engineering, Sun
Yat-sen University, Guangzhou, China. His research interests include wireless
powered communications, Internet of Things (IoT), low-power backscatter
communications, and machine learning in wireless communications. He was
a recipient of the Best Paper Award on MAC and Cross-layer Design in IEEE
WCNC 2019. He has been the Lead Guest Editor of the IEEE TRANSAC-
TIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING, a
special issue on Deep Reinforcement Learning on Future Wireless Commu-
nication Networks.
Xiao Lu received the Ph.D. degree in the University
of Alberta, Canada, the M.Eng. degree in computer
engineering from Nanyang Technological University,
and the B.Eng. degree in communication engineering
from Beijing University of Posts and Telecommu-
nications. His current research interests are in the
area of stochastic modeling and analysis of wireless
communications systems.
Dinh Thai Hoang (M’16) is currently a faculty
member at the School of Electrical and Data En-
gineering, University of Technology Sydney, Aus-
tralia. He received his Ph.D. in Computer Science
and Engineering from the Nanyang Technological
University, Singapore, in 2016. His research interests
include emerging topics in wireless communications
and networking such as ambient backscatter commu-
nications, vehicular communications, cybersecurity,
IoT, and 5G networks. He is an Editor of IEEE Wire-
less Communications Letters and IEEE Transactions
on Cognitive Communications and Networking.
Dusit Niyato (M’09-F’17) is currently a professor
in the School of Computer Science and Engineering,
at Nanyang Technological University, Singapore. He
received B.Eng. from King Mongkuts Institute of
Technology Ladkrabang (KMITL), Thailand in 1999
and Ph.D. in Electrical and Computer Engineering
from the University of Manitoba, Canada in 2008.
His research interests are in the area of energy
harvesting for wireless communication, Internet of
Things (IoT) and sensor networks.
1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.
32
Lei Shu (M’07-SM’15) received the B.S. degree in
computer science from South Central University for
Nationalities, China, in 2002, the M.S. degree in
computer engineering from Kyung Hee University,
South Korea, in 2005, and the Ph.D. degree from
the Digital Enterprise Research Institute, National
University of Ireland, Galway, Ireland, in 2010.
Until 2012, he was a Specially Assigned Researcher
with the Department of Multimedia Engineering,
Graduate School of Information Science and Tech-
nology, Osaka University, Japan. He is currently a
Distinguished Professor with Nanjing Agricultural University, China, and a
Lincoln Professor with the University of Lincoln, U.K. He is also the Director
of the NAU-Lincoln Joint Research Center of Intelligent Engineering. He
has published more than 400 articles in related conferences, journals, and
books in the areas of sensor networks and Internet of Things. His current
H-index is 55 and i10-index is 210 in Google Scholar Citation. His current
research interests include wireless sensor networks and the Internet of Things.
He has also served as a TPC member for more than 150 conferences,
such as ICDCS, DCOSS, MASS, ICC, GLOBECOM, ICCCN, WCNC, and
ISCC. He was a recipient of the 2014 Top Level Talents in Sailing Plan
of Guangdong Province, China, the 2015 Outstanding Young Professor of
Guangdong Province, and the GLOBECOM 2010, ICC 2013, ComManTel
2014, WICON 2016, SigTelCom 2017 Best Paper Awards, the 2017, 2018,
and 2020 IEEE Systems Journal Best Paper Awards, the 2017 Journal of
Network and Computer Applications Best Research Paper Award, and the
Outstanding Associate Editor Award of 2017, and the 2018 IEEE Access.
Dong In Kim (F’19) received the Ph.D. degree
in electrical engineering from the University of
Southern California, Los Angeles, CA, USA, in
1990. He was a Tenured Professor with the School
of Engineering Science, Simon Fraser University,
Burnaby, BC, Canada. Since 2007, he has been
an SKKU-Fellowship Professor with the College
of Information and Communication Engineering,
Sungkyunkwan University (SKKU), Suwon, South
Korea. Dr. Kim was a recipient of the 2019 IEEE
Communications Society Joseph LoCicero Award
for Exemplary Service to Publications. He has been a first recipient of the
NRF of Korea Engineering Research Center in Wireless Communications
for RF Energy Harvesting since 2014. From 2001 to 2019, he served as
an Editor and an Editor-at-Large of Wireless Communication I for the
IEEE TRANSACTIONS ON COMMUNICATIONS. From 2002 to 2011,
he also served as an Editor and a Founding Area Editor of Cross-Layer
Design and Optimization for the IEEE TRANSACTIONS ON WIRELESS
COMMUNICATIONS. From 2008 to 2011, he was the Co-Editor-in-Chief of
the IEEE/KICS JOURNAL OF COMMUNICATIONS AND NETWORKS.
He served as the Founding Editor-in-Chief for the IEEE WIRELESS COM-
MUNICATIONS LETTERS, from 2012 to 2015. He is an Executive Chair of
IEEE ICC 2022, Seoul. He is a Fellow of the Korean Academy of Science
and Technology and a member of the National Academy of Engineering of
Korea.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2020.3004197, IEEE
Communications Surveys & Tutorials

33

Ying-Chang Liang (F’14) is currently a Profes-

sor with the University of Electronic Science and
Technology of China, China, where he leads the
Center for Intelligent Networking and Communi-
cations and serves as the Deputy Director of the
Artificial Intelligence Research Institute. He was a
Professor with The University of Sydney, Australia,
a Principal Scientist and Technical Advisor with
the Institute for Infocomm Research, Singapore, and
a Visiting Scholar with Stanford University, USA.
His research interests include wireless networking
and communications, cognitive radio, symbiotic networks, dynamic spectrum
access, the Internet-of-Things, artificial intelligence, and machine learning
techniques. Dr. Liang has been recognized by Thomson Reuters (now Clarivate
Analytics) as a Highly Cited Researcher since 2014. He received the Pres-
tigious Engineering Achievement Award from The Institution of Engineers,
Singapore, in 2007, the Outstanding Contribution Appreciation Award from
the IEEE Standards Association, in 2011, and the Recognition Award from the
IEEE Communications Society Technical Committee on Cognitive Networks,
in 2018. He is the recipient of numerous paper awards, including the IEEE
Jack Neubauer Memorial Award, in 2014, and the IEEE Communications
Society APB Outstanding Paper Award, in 2012. He is a Fellow of IEEE, and
a foreign member of Academia Europaea.
He is the Founding Editor-in-Chief of the IEEE JOURNAL ON SELECT-
ED AREAS IN COMMUNICATIONS: COGNITIVE RADIO SERIES, and
the Key Founder and now the Editor-in-Chief of the IEEE TRANSACTIONS
ON COGNITIVE COMMUNICATIONS AND NETWORKING. He is also
serving as an Associate Editor-in-Chief for China Communications. He
served as a Guest/Associate Editor of the IEEE TRANSACTIONS ON
WIRELESS COMMUNICATIONS, the IEEE JOURNAL OF SELECTED
AREAS IN COMMUNICATIONS, the IEEE Signal Processing Magazine,
the IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, and the
IEEE TRANSACTIONS ON SIGNAL AND INFORMATION PROCESSING
OVER NETWORK. He was also an Associate Editor-in-Chief of the World
Scientific Journal on Random Matrices: Theory and Applications. He was a
Distinguished Lecturer of the IEEE Communications Society and the IEEE
Vehicular Technology Society. He was the Chair of the IEEE Communications
Society Technical Committee on Cognitive Networks, and served as the TPC
Chair and Executive Co-Chair of the IEEE Globecom’17.

1553-877X (c) 2020 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
Authorized licensed use limited to: University of Exeter. Downloaded on June 24,2020 at 08:08:28 UTC from IEEE Xplore. Restrictions apply.